From da4bcfdca6592278a22aff2f86e59e555f3ea08b Mon Sep 17 00:00:00 2001
From: "lea.douchet_ird.fr" <ldouchet@hotmail.fr>
Date: Mon, 28 Nov 2022 13:17:02 +0700
Subject: [PATCH] install.packages stat
---
07-basic_statistics.qmd | 34 +++++----------
img/dist_filter_1.png | Bin 6761 -> 2752 bytes
img/dist_filter_2.png | Bin 14430 -> 5465 bytes
public/07-basic_statistics.html | 41 ++++++++++--------
.../figure-html/LocalMoransI_plt-1.png | Bin 44322 -> 44265 bytes
.../figure-html/MoransI-1.png | Bin 17446 -> 15732 bytes
.../figure-html/kd_test-1.png | Bin 15736 -> 16007 bytes
public/search.json | 2 +-
8 files changed, 34 insertions(+), 43 deletions(-)
diff --git a/07-basic_statistics.qmd b/07-basic_statistics.qmd
index 2503c46..4490349 100644
--- a/07-basic_statistics.qmd
+++ b/07-basic_statistics.qmd
@@ -4,7 +4,7 @@ bibliography: references.bib
# Basic statistics for spatial analysis
-This section aims at providing some basic statistical tools to study the spatial distribution of epidemiological data. If you wish to go further into spatial statistics applied to epidemiology and their limitations you can consult the tutorial "[Spatial Epidemiology](https://mkram01.github.io/EPI563-SpatialEPI/index.html)" from M. Kramer from which the statistical analysis of this section was adapted.
+This section aims at providing some basic statistical tools to study the spatial distribution of epidemiological data. If you wish to go further into spatial statistics applied to epidemiology and their limitations you can consult the tutorial "[Spatial Epidemiology](https://mkram01.github.io/EPI563-SpatialEPI/index.html)" from M. Kramer from which the statistical analysis of this section was adapted.
## Import and visualize epidemiological data
@@ -191,7 +191,8 @@ Under the Moran's test, the statistics hypotheses are:
We will compute the Moran's statistics using `spdep`[@spdep] and `Dcluster`[@DCluster] packages. `spdep` package provides a collection of functions to analyze spatial correlations of polygons and works with sp objects. In this example, we use `poly2nb()` and `nb2listw()`. These functions respectively detect the neighboring polygons and assign weight corresponding to $1/\#\ of\ neighbors$. `Dcluster` package provides a set of functions for the detection of spatial clusters of disease using count data.
```{r MoransI, eval = TRUE, echo = TRUE, nm = TRUE, fig.width=8, class.output="code-out", warning=FALSE, message=FALSE}
-
+#install.packages("spdep")
+#install.packages("DCluster")
library(spdep) # Functions for creating spatial weight, spatial analysis
library(DCluster) # Package with functions for spatial cluster analysis
@@ -225,8 +226,7 @@ For each district $i$, the Local Moran's I statistics is:
$$I_i = \frac{(Y_i-\bar{Y})}{\sum_{i=1}^N(Y_i-\bar{Y})^2}\sum_{j=1}^Nw_{ij}(Y_j - \bar{Y}) \text{ with } I = \sum_{i=1}^NI_i/N$$
:::
-The `localmoran()`function from the package `spdep` treats the variable of interest as if it was normally distributed. In some cases, this assumption could be reasonable for incidence rate, especially when the areal units of analysis have sufficiently large population count suggesting that the values have similar level of variances. Unfortunately, the local Moran’s test has not been implemented for Poisson distribution (population not large enough in some districts) in `spdep` package. However, Bivand **et al.** [@bivand2008applied] provided some code to manually perform the analysis using Poisson distribution and this code was further implemented in the course "[Spatial Epidemiology](https://mkram01.github.io/EPI563-SpatialEPI/index.html)â€.
-
+The `localmoran()`function from the package `spdep` treats the variable of interest as if it was normally distributed. In some cases, this assumption could be reasonable for incidence rate, especially when the areal units of analysis have sufficiently large population count suggesting that the values have similar level of variances. Unfortunately, the local Moran's test has not been implemented for Poisson distribution (population not large enough in some districts) in `spdep` package. However, Bivand **et al.** [@bivand2008applied] provided some code to manually perform the analysis using Poisson distribution and this code was further implemented in the course "[Spatial Epidemiology](https://mkram01.github.io/EPI563-SpatialEPI/index.html)".
```{r LocalMoransI, eval = TRUE, echo = TRUE, nm = TRUE, fig.width=8, class.output="code-out", warning=FALSE, message=FALSE}
@@ -274,16 +274,15 @@ Briefly, the process consist on 1) computing the I statistics for the observed d
A conventional way of plotting these results is to classify the districts into 5 classes based on local Moran's I output. The classification of cluster that are significantly autocorrelated to their neighbors is performed based on a comparison of the scaled incidence in the district compared to the scaled weighted averaged incidence of it neighboring districts (computed with `lag.listw()`):
-- Districts that have higher-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local $I_i$ statistic are defined as __High-High__ (hotspot of the disease)
+- Districts that have higher-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local $I_i$ statistic are defined as **High-High** (hotspot of the disease)
-- Districts that have lower-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local $I_i$ statistic are defined as __Low-Low__ (cold spot of the disease).
+- Districts that have lower-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local $I_i$ statistic are defined as **Low-Low** (cold spot of the disease).
-- Districts that have higher-than-average rates in the index regions and lower-than-average rates in their neighbors, and showing statistically significant negative values for the local $I_i$ statistic are defined as __High-Low__(outlier with high incidence in an area with low incidence).
+- Districts that have higher-than-average rates in the index regions and lower-than-average rates in their neighbors, and showing statistically significant negative values for the local $I_i$ statistic are defined as **High-Low**(outlier with high incidence in an area with low incidence).
-- Districts that have lower-than-average rates in the index regions and higher-than-average rates in their neighbors, and showing statistically significant negative values for the local $I_i$ statistic are defined as __Low-High__ (outlier of low incidence in area with high incidence).
-
-- Districts with non-significant values for the $I_i$ statistic are defined as __Non-significant__.
+- Districts that have lower-than-average rates in the index regions and higher-than-average rates in their neighbors, and showing statistically significant negative values for the local $I_i$ statistic are defined as **Low-High** (outlier of low incidence in area with high incidence).
+- Districts with non-significant values for the $I_i$ statistic are defined as **Non-significant**.
```{r LocalMoransI_plt, eval = TRUE, echo = TRUE, nm = TRUE, fig.width=8, class.output="code-out", warning=FALSE, message=FALSE}
@@ -319,13 +318,11 @@ mf_layout(title = "Cluster using Local Moran's I statistic")
```
-
-
### Spatial scan statistics
While Moran's indices focus on testing for autocorrelation between neighboring polygons (under the null assumption of spatial independence), the spatial scan statistic aims at identifying an abnormal higher risk in a given region compared to the risk outside of this region (under the null assumption of homogeneous distribution). The conception of a cluster is therefore different between the two methods.
-The function `kulldorff` from the package `SpatialEpi` [@SpatialEpi] is a simple tool to implement spatial-only scan statistics.
+The function `kulldorff` from the package `SpatialEpi` [@SpatialEpi] is a simple tool to implement spatial-only scan statistics.
::: callout-note
##### Kulldorf test
@@ -335,10 +332,8 @@ Under the kulldorff test, the statistics hypotheses are:
- **H0**: the risk is constant over the area, i.e., there is a spatial homogeneity of the incidence.
- **H1**: a particular window have higher incidence than the rest of the area , i.e., there is a spatial heterogeneity of incidence.
-
:::
-
Briefly, the kulldorff scan statistics scan the area for clusters using several steps:
1. It create a circular window of observation by defining a single location and an associated radius of the windows varying from 0 to a large number that depends on population distribution (largest radius could include 50% of the population).
@@ -349,13 +344,12 @@ Briefly, the kulldorff scan statistics scan the area for clusters using several
4. These 3 steps are repeated for each location and each possible windows-radii.
-
While we test the significance of a large number of observation windows, one can raise concern about multiple testing and Type I error. This approach however suggest that we are not interest in a set of signifiant cluster but only in a most-likely cluster. This **a priori** restriction eliminate concern for multpile comparison since the test is simplified to a statistically significance of one single most-likely cluster.
Because we tested all-possible locations and window-radius, we can also choose to look at secondary clusters. In this case, you should keep in mind that increasing the number of secondary cluster you select, increases the risk for Type I error.
```{r spatialEpi, eval = TRUE, echo = TRUE, nm = TRUE, class.output="code-out", warning=FALSE, message=FALSE}
-
+#install.packages("SpatialEpi")
library("SpatialEpi")
```
@@ -387,7 +381,6 @@ kd_Wfever <- kulldorff(district_xy,
The function plot the histogram of the distribution of log-likelihood ratio simulated under the null hypothesis that is estimated based on Monte Carlo simulations. The observed value of the most significant cluster identified from all possible scans is compared to the distribution to determine significance. All outputs are saved into an R object, here called `kd_Wfever`. Unfortunately, the package did not develop any summary and visualization of the results but we can explore the output object.
-
```{r kd_outputs, eval = TRUE, echo = TRUE, nm = TRUE, fig.width=6, class.output="code-out", warning=FALSE, message=FALSE}
names(kd_Wfever)
@@ -465,8 +458,3 @@ In this example, the expected number of cases was defined using the population c
In addition, this cluster analysis was performed solely using the spatial scan but you should keep in mind that this method of cluster detection can be implemented for spatio-temporal data as well where the cluster definition is an abnormal number of cases in a delimited spatial area and during a given period of time. The windows of observation are therefore defined for a different center, radius and time-period. You should look at the function `scan_ep_poisson()` function in the package `scanstatistic` [@scanstatistics] for this analysis.
:::
-
-
-
-
-
diff --git a/img/dist_filter_1.png b/img/dist_filter_1.png
index e11d04497a5dd79cb75c5f49140e6855f14397a7..72efd1a6323904b0bad518a240ff8d58315d1ac5 100644
GIT binary patch
literal 2752
zcma)8e>ha>9)Hg{X2=YMWo9F;tJ>>lQz|1$?xsa4`$NoQjB!OPW{rf=4ENYgl9s)-
z%PteGmR3K^l#Qv;bYjbH6s0HR*O5gEGYW$t_nq4N-+k_L&-0x3ocHs7zTflteBbYR
zo;Npm&qpT4D~$mFCIS9CJ^=tB^|uiQkwD`;=iwtdc=u<1?)r=4?k;xcNZrLs_{QN#
zInpk%9ukW=Qn6Ai<+w{ZVriFFDRx)FQL9vHyOdgOOG`_al+)EE?dsCQU0N+<XuG6u
z7vZ`dSXx?A{@@Z0fH5;*hwo=1>3DDMsi!;N+c}Wm6kyZc<#E70bLsra?Z-pfUOX8e
z760?hr&a!6QPxZb(U|uBk^VOxizz<~s<g(z-Jhr}&(>D-Id2#2x*3W;Ej%Iv*Ia|z
z_oeh~o=4@9xMK7Yr??Wt7gjnF#T9W|re1Y?@myQ9SABPB_>O9Rpy_o_devah=$CPo
zgS8KftjS%MV~CNw#3IAakK&4XCJjY<`xkHM)V(E(OOKk%e1dA?zF*L_XQfEHBZtN(
z6wh^0x3!Iw#)T1e(-1NdQC#Uk<d`iCXs5>e8%xD2Rvpluu2W(fwuj5~@JL!t#j%TY
zn(669B2(Auagxt~lD1NpGN_rao0~Rj#;Xcy#4Slp5~-J4O5eq@)QPe3jBhJPTa0W|
zuPJ*+3f2ws(u#xZGNpx!!cM>Cw-{wUJalPt6Y)i0`(e}Bk#>v0!Lc5rzQ{K-WwQ5g
zZ`;*Abvs()UzA^yTyR^LC&RV3kacB!*jV|-pv0_O^8=d4qMkY1(v8b&8qc07^(#ou
ziTrA@=J~)#qx<v0Hw&+u7N?Pk$l}V&FBF0)3V-g=MoL)CVd;dwGkymP^ZtJZbzqft
zWf)bH;#koS3I%Pw=U3A;Rfqqegk3%DMx2Ovg}QWXk`=NDZ9SlU7b`-n9PhtGFSSK;
z*AEC#pF0$`#FMus3P%d#vt;=iY8GSeo)PBaR=<_3P=Tg|E3)G(;#osD<~I4=3ThZv
z6isBS&<2o4OEhMSPx=Onb`hKE#RF}>b|cj*Rg<|M7`N5BMC3p?A5-t>5nDKQQ#A*9
z_}0{?vX!&PEDNc&wp&i$nee<SXK5yP`#SXf-0+GxhKxBUlixJlAZAqd*$9>Wxhswt
zH=`btl+L}N0C91nvxxQa1~z+`d_0<}VM0j&v#E(CZXbp0SWr-f*1a}a3t49Iqt?41
z%i)O?`;>gGFgg+!H-T+ymYr04;z&ND^~Y96bWm(YJXM__=wvf^GDBgKpAU#FAp&!{
zNMsn33?2|$5FqD*z0f~$6Q8OPz!k?gCsRtSSca!2P$sz@kuV6*l&e+t8?uNbbaop|
zZb4`rk-8a*A{n$A$ZOX@!ZPv)KxOX+tv?yG{IKYb(1lW<TFw_l>gnR=`o+L{dMWK4
zVlIB(1(BVz4`9;usA!udUt(v#BJ9yw-@g*|Yvw<OI?fKOfYyY}fw^#zQRepsENnIj
zU}HWrZ3P&Uuf39!Y5vyBm=9hCWynxS;DC4?N|)={F+&{4-TL6=Kle$XvG%wA`RYkQ
z>RuS!4})#r1$*OVhVt4(5uYktizBJ4WT!SDboonm;kV29_|_&br(`%C0S_oFq7-Ap
zVu(#i?(Hnk#5Ig}3A{`<??;*ZjwHgzLXYI>b{fWl5j3ou!B2wzQ&$n!0ktz3LQH<s
zq?4mH0le~k_QyaSO-XXr+o?8rzkWJddM^RWW<QAkjmjeO2YYG~fY0F9YMREqzpvzg
zem7x$)MbJO)eO3Fx}dFIH(}u3>Pv&lD01Lw4oL0ULJkUA9w(Qa#KP+Dqltb@{^VwQ
zy~)5hYR7Y&?!#vU?~67Tiu)wJceD1k-K_hGb0y2{0sP*d{vJ$-e!L(Q@{bAK-l~V2
z*O}tod=G9=HO7AH(JnnC1F_8zC+<}03mX=}BtH%-WRkfGZQ({IwtQeTyIlYCz(35!
zx^vb*=of|Gfeqi8IAizl#}8otX2xsmy$&_}D=S=b+?>cqMZVDYwas?P=k?vq!$bja
zz6mcfkSkcQ<|gDCxD6eEXa46zO3BZ7f)Qy%$5~zC`0?E~48>4g>6vX&qWMn>6~6+X
z7}Y|RmfB%yrMRv>lcsj9rty;oekNqFmKUEO>`GxUX{!Ra{t<zHXpGV2jq|hU!a<~-
zjUzLmA4KAj&Ug`Q&Nrwi-7HHyhb{yn+!X{+(#S7SH0`yCBUIUkDpo`KzK=-!{MKKw
ziBYkGL^SwIaumKR`-8Uj?q?Y^jLY3ZRyLQdG8Cqd##(Y|IoO0+Ty`J|cRl1nmj`5K
zpF?z0DeyH2SPxltUS<*dJS|PBV;vnASw!k}Zw-*gnC3<!UfRX`_Unl!o(|N`ZJ`aI
z+fXQ2FF~r^jO5NyxV3>W1{4_zzXH|HCu9%^HxPz^Py=BesCLX>M}#0;3&Isuti|I|
zML_0J<O_4c65)PoL;9=qv=HF2jC=*EjmfnLm$#blqtXK>K(!NHxH=MdMip+j^&h`6
zNGpVEMx^*5ck4Fb<Cwx*5|4FmiB$RG*c{6J^e_AaDquZw--F6hO(O{_iI8k$W`rh>
z1EH388RQMXGb4_9PEk5H;4e{;pBdh0gmF1^{(&-sOgjIZ<Q1CmZ72_S4a`hOc;SaG
zD=|gw;Y-?x(<bDFg~mMy*OI8bj^z*1py2Ld6ZSXjh)qa>)LrkQh;ferEo^SJ6uR`U
zAZDwL$OcfmoP;)+#7sJxKTH!w0S_ZRU`3d~&x0ZgEEtsJm;IWeaQTnH`n7l-s_+G}
z-+kn;5v&a~7l4~FUk7-WgddV%M_fb|f5Y!557FcqKt?0W!RF;;3c`Jl7)I*dV646%
zpu|vfPapOY@ZQ&lN!=sy%a}+27!-LDkhyLU2FfOu%LQ88;>D!Se0MgZhe_1-_6<3_
z7%GZlnm+r5?4a<wtJRZpYW5X#-qjC}k6168eRwn8!MpEBYz~kM@+5+z?d8^3XYDT<
zRfKN4WSSB>I&b&r(RadP1I<8g=&G@nj-rZW4`;JB54!M;DyXhefXL?pXPe!H_anlO
za*b>rdKd}2B9viFQWesFjh4;mJLnYN!^zJIrJspW#^Jg`WBXxQ0@bEH)Jk;Z3MLOH
zQ&eenS5Y?7Q@(%0Av)gVbIxEoh3<W`G9tZoo0;47p*aT*@`_%cxVTCe%y!~%`m#)p
ztftHTK_2E~P7SNzkjisM%LT~nGb8n~H0vA6uR&Rf7a{Ati`OxTXNZgK{~Y>US-J0y
pXwy98VEBuh=;PY|c9zrG1=MP1VyAwW!6PXM@Y}Pa`tSTN{|(+#^#lL_
literal 6761
zcmZ{pc{o)6`~T4lW+vX&q3qdZC=wZwEo;cW#h|i9V`<3l$R5g;eP6RQgAiGcy|Swe
z!j#H37?jBNoA>8)eSg>Q`u(ooALlxc*Zn-#ea^Y>``p*-x=*}`kq#65Je-DxhDlFX
z%an$Omh|tzc!p}xhZ|MW(9m5lF*MhfmzTeC<%)uW!qux+6%`egl$5SryLSEhbyZbW
zH8r&xH*Tn_t7~a#X=`iWym?bcM@LszS5HsRz`(%J(9p=p$i&3N)YR0>%*@=}+}hgO
z#>U3h*4EC>&febs&Ye3B4i1ivju;Ha$;rvt+1bU#<?h|P9v&Xv-rf%%KEz_NK0ZFa
zzP^5be*XUcj~+e3;cx)~0fB*mj~_n{3JMAi4h{(k2@MSm3kwSm508k5h>VN`03a$V
z>dBKQ(b3T{F)^{Rv2k&6@$vBq2?>dbiBF$CO-f2iPEJlqNqP3{+4JYmQ&UsZ($dn?
z(=#$MGBYz@ym*n7m6e^Hos*N3o12@LmzSTPUr<m`SXfw8R8(ACTvAd}T3T9GR#skK
zUQtm|Sy@?CRaISG{qp6@nwpy0+S<Chy88P1SFc{-@puA((9qD(*x1<A)b#rG>*nU>
zmX?;**48&~-n6x~wYRswefzefqocF4v#YDCySp0%!JeL;-rnB6zP|qc{(*sk!NI|y
zp`my0-o1bS{=<h4!^6WPBO{}uqhn)Z<KyEK6BCn@lT%YuA3uKl^y$;*&!4BKr-?-3
z%*@Q}?Cjj!+?Ow3=I7@Z78Vv47fB@2*RNl{efzewwDkS^_a8rg{QUWId3kwdWo31B
zb!}~JeSMuwCU0zPY;JCDZEaB~l<n>9ot>TC-Q8cme*OOadv9;=&!0aB2M2%u{yjWA
zJUTi$Jw0`CEu_{AH=UlAx_MyUM*gAWN2{R0${8{DMz4UD4ex2(+&nHI*F)sOB?lUH
zdhYuJm%fTgT=HdQ(YoOcFbDj+tI#&l$KAjtnvfHhkhZaRy8fj#Nxt`wFMq+Vir<2o
zh~`f8^dG;4R|gKGKdvR8PT>b4L+Q?2H(`Fdfx{C(b}3X|2XTKi2E)KnwR*+&Qow(e
z8Oi;Lzcp5`_)iBP4R+cw4{Dvi7k=nAS2;w0eTx-8t%VAwi3B?SDh%J<x(vQ*)(aKS
zCs8%$*sio+jmW3Bq`*_;A$>};TJT|jq5yN}GEvbkpqKS@1&flk=!XufX(xx)EKQW=
ztbFZC;N?1)@mcwhksMIcRS#WCbHIMTxW6QR`rC3U%VjZOv&$&t=(yLhGZ-bNv@PjN
z26MNLr=|*Zm{a(<4h|eG!?puTG{U1k3HD##r&N$i)nk_`*S9w9LOqvUy0@@Lu7=oe
zX5Wd6vd4RMMA2(O_TODkZDN;RbWH%C1eX+vrz?$&r}(XAm!+k9mA%_(S)@Mdi-!L=
z^mzSyqlz>~SiJ67u?4O@yN&gs>ZOZHd!avx`^57H3!MV&dHWpK_FTia_WSGdsz_tb
zU3LM=JFlW_@#pWd9gU7zh8^QJ3=QqSEp7219p!I7EM*W?)lRNEd-A<weRSw63_Sfi
zqIMcstg_RHtfe^zNs;L~KFJq1JLP64$A$Z+%gS=f_9rKWC-5F_5~nhn1+<w@E$s12
zE0d+0Ev5ektTebG>%WOiMh@pO(S{#5GF?1L71@u)$8f5-fp7nJqNzGBD(jl<CBOfx
z0{%y!@YIIJhSJMT0BL~}&&^UonW-2+X$S~Usp?imWV?ZL)&hTv>=<t!F&=7(lt3Rg
z%pN$U#7)k{hDxzb086!ZWMs))J{o!T6Smm&U<X%M-3O^H$)OrM<3R6MyUmg!qEYXK
zJ;`^P2nTG&SYp26A;DOm-$rJCAzRZ`;~SGNVa$MkMtUD-7Ir|)#R<2bJPGxl9I^#l
zc$nM7%x04=eT)nZnjA6$|AfSPR!+?%nI?qBOb+RTH4Lc&HL31z;zE-fhHirEqE-Rh
zJdZgY22>67?v`Rs`|lhlWXq8o)3uQBPV1I-M<Kcn;<98cnLVWOW1OQ9eAA_{nxnyp
zl9tE8$5Wjp1FBj`-pwav3b}+uyLtAgTAjO_OZu&O^B6ucTx_5`2T6I&%_Tjn@`^Av
zeRW2xwU5N7WYm&)G=$BBbuL|9vWheKAxkix5t|WJ^2ZG1VJ@4r#e`1o4T*tc$NXko
zE8N1-j!uOws0Lm9exdqQknk|4#r0m*M?80vNDASrC7fNX1Nl#uC+;VbjQ9Sr`^fWy
z9N$i)txe?vrah$U!a|pZxeQ%g@w%3bosYUF=K~lxH;9t8pbcdzSGMFdXoe(rN;wis
z3QL{0&4W(wr~I^1VCGv{oG)OV)gA3~kq$(uy%Z(7Z*8_G33F6#UqXJF7XI)6$aW^{
zA~Djfh7F5DbCZ{FV!IU1qg9<rXrg^ksBH+eJs|h+L8L>_RQU|t0hPXG#i)prp2Bm6
z;2;|Kv@awfJMaxlnD(3!PNtD+g=k@`;OG}NO&Gst7Z;16xurh*=qx?za`<*|@@L?d
zuQLpteXR+TW_W;4%hp6(HDIl~EQ%sZgC1nJ3X;_RHcSF6!9vk_=UU;Loe;6WtdC^r
zdf*NwELvia*_FNmD2#k6Dk@RopLAcU)p5^1f8#99c~G9#4E!M{_P!!|QX!|3w3VXq
zms7ijNSoAaM$u)cA~|R%GzG$&->|tW!n?Ye-FP#B>Aa$_IT;MY6<Pz(Wm)jps13iH
z9a&CFduFv_!d$E)T>!igY$s@GGolM6mEd~}0n>N0CsGJY*-^!${g)aN%&PsPuFNsO
zVH@o+Jw1LY`$;h->UR-}5D3i5g@*wT3+On=ZSj0#(npYX*RMxLVNdc$uaF^XXy#=r
zUa{YB-l%+1#Mgw)pVt<n@<&z3kltFUrvz(5FDKKiLedPq5?El2pR$R_#Z1#>Ef#U1
z^Ek6$Vb*7VS}Aq)8@WR71I8)6u+d&-SJ-{vpWXw`3`dvBMmkyW!;c8yP%jg@EZU0j
z1`}9>j-N#{f9J|txs{bcdNlqV^Yt%1x(I3w>vT9H)%s%A8p~k`@QA5HWli708rkfj
zYcGEOdnp|T92~NEc~^`Zryj)uCDNk0^(e~Z91S5ABJ5vQ<iv+*_UYg4CyDIeyk1LU
zO*mY`vW1khTT-&2<R?gyNGk%XY?;m$WVRUyQ+?<fb^DA-NZftV{nWDS<Vp2yp_~j7
zLB^y@M4-e5RsfHHiyg3x=wqu`%_;7YeB<yu$RfaplD^O^r^GBy?E^)?E@9UONFy1{
z7Xf;zT8(TWY#+DY3ZM%`xv5_wl=F-H6TWe_r;uRR8_>ePyyzm)eZ~{!z}CeBIJ%h1
zn>3qMA~WnGdX!jv*y;xQd~6cQ1XWAg?IptMz?;|u-dfU<F`nHvDv>n!tp8DoxwCi^
zBNynZgE@Ba>N9d<6$2avfZrF7zUzMx#a!68Wyh1~JQSo9F&A3T`WK3*Ka!9f)}|=p
zm(H@rbwd4E?Gf$X=5Z_|gq}<6;48tz^}g20r}-07(vUM_pW?CJZ0J%8YexRHXk^IK
z?;*7k(vay}nh2bWPSaTgh&Q`-0kV!4wvCC#tYjRQ6nsx#;g*I7L?Z2P_+8cEU<L#L
z?|WQA=8=38%Dj*<g)<0mf&OGSK4t(NF%3=Zl_dz<#ztWT$YG1lFVCZknUom!y8m((
z%n@N;-RhV8BhjYlF!A0S6qliS31NG3kC7V#J>WXl10OwPMi=vOGcwa`HnBOC@S}@R
z+>G@=e!bZ*mj7@}=s=Dz+D4rD42aicDzGCSkkuV2$NIWmLX%s%PRxM#tRPeoO4yxw
z(n%T+EF=9>Z4|f|(TPzkTT}&t!?vDuV#Y`Ca7OQ50{afoMPIjMe8eX?e1p;rz0vhz
z$gy}8v8TaI-@fTG*vRx4ENa0tc*0wXWCAz1ix#8Jh}&e%JCqA|lU54U$$@*tGOaM*
zuPYq$P^MfHuCucv>b~t0B2|c(`Q5`lI8Jqrh3o9xh>*-L7ltM#AFybiSL4Zw49|Q-
zD`wm~nep*D5w@XWe~FB;j?3YpatYM6%!LkJd?yYO<M4XYsCoZAy)7PDEBDCE7JEMq
zvAHz`$ZOhT8C%~!fR1$dH>=xI<Q^DXni$Wh-{))Y9wjgkX}vV59uq0pGdPt)7OtZ*
zZd0txY6j>NhdANzZ7FL!E9P&)r2<jPxnhx!Z$#e*Riu$Id|#?ID<CGzjjxfr6v1Z%
z?}J)J9^@;>=ig%PgZ6faDvUX~rk(KuJ+WwHflVK?IKwQUrdAH!qVoc+W?z%;6roz7
z_4GN&S~~vL^KQ0ycz1_re5WcI>r2g{uXKD+WG$o&8kJ0M6%x;945uRU4dS!&U>cn}
zh;XH$HvgRumL=oJvu%Y&hyx3z+q#I(1sch~=lw27cd$ip*-wkV(c76JoyHw>G>?Ki
zNLM(-Uoo_&uu&$<Ea}r4bP-`Du#48bWomV)O{7n0-9W7SRx4X{6UkM(8!%RI)@fpL
z2W7vC`D<k#Vd>Gup}N#O{XKl@$%!t6xFKS`+^(!XgU;t@B3*g9kMDh_jxZYeayUl#
z58XPkBF>1Y+u^z1Kipu%U9oWkt5=pU69q&)p)V|68<!o1%WJ>Fr|AiMLYHOlc<lwk
zlPy}3L^-c8G5;K2kPBpJf*#h6CCViIaaQS)XQdp3@H=)+*^vB|I~*RmC?3q}%~3<>
zJ<vc(zD!PA<Z(kVKea)sEDx$4bVN3j77Q|>^2E$>mL_Nl&W6i=qw_D@;n5$OSzwtw
zoO|cNJ3nL9-eejkFEAkVDny5`1<9p^P+3UX#E8>IHIrOZ;qxO_|7C0>FZfL<mT;Wz
z240RCbrO#RS~2dWOxK8dMS?Qcv%`N<l%&n{pg!PL<<bDT!Mib|JS<#ae6>?C&&Qk<
zD7h?LA#ZdtNogsB#_k*T_`3R9zP!J5XvEn*X!xXr0oIK1KfIXU40)Tjx_`+BEHtpA
zQ~el~52gPGl%r_>`7sZU(w*@E|GB#nf(xNV!&&;EuQK0n+i8cfzxSsWZAad=hFS`f
zlBv2qKF=cqHA1{bqCVRw8zug@h2R%Kt$V*~+Puh=#*UlpTq<IoQ!#%%t`mzcc)Kec
z2hX2<KU*HrOnMYAy4W6=%s7QqnG-R_Gmgt9WYEn5p}TC*#0n%?g=QQXVqM`VbaYlH
zomy_FVApmzveu3es?}z1z%}DYSw$C7pH8+iv~bZ$WxfZFrMyE*`%dFE-^DbLYG_vh
zbj*JkvzX@(maw7IA4niZ)3&TQeC_V*T2p+lMUWz8$kwSs()-Q=5t!S_d<Jq)po*0!
zZMRtG`_pbcz_BU1L{aG&hLd^2f;?;HvFEHtMK@4yc&A!~R}*0rV*^#`0lYpB>Blml
zU)VR1?j~cFQxlPSPuc%nVo=r=O&DP}@Exzgc{T?D)Ot{4nhE_AXuA`nJ?P=X8sUz~
zKna$$5vGuo7gPNHP}z)b#)9>20TgB5?(mg^Biw?cUy-TRzjaj-hw&WeBxlF4^17br
zhX;35de^iEw{oEI3PbSDe||xY3_ev7pQF?*u@=}A=-33WWQEi|7=L~5qSnkE^0;M@
z6fwYS3_A9VTrMWHuRc+5$XT+&Lna~xrCK+o0FqupZaFJwULwgVPK9bMlH9LvGCip%
zBHp@XhAnN<*s7p7FE`WN0#mudJi!DXM@yI~%KBE0&h4WZiO*k!B)Oe$r_ur(A1Edq
zZ!c8G;B92g?6Yp(#vTeLlL|f%-=S{YzBDPEnN`xd5?Msz(s>R^vp+F5p}cMCNF6ra
zhoIXZGs5fV){9SWDc>YC=EfFbie}R3twr`P7#dRyBrT5W)UitO1tgYPYD0zxHR4ua
zs!2Em()lN8l<uEJ=0nviO%H~sarUPHh-4P2b)r(rCt72NcL&<fOZNtPze8LWfYPQ+
z9V<euai%~Q%cATB>zBXo|MMMkT1;z&=SE|<l25U_pS=Y)Hi%xSFkc{7ob|^*mTYa#
z?6!cux3i)?{Sd5PV=^bQ|9A{<Cwl!m#Ndmr&iN%yVqGe~2e3tlI&9XdY%`;+f^G75
zXJ)4k(BOK@qH;)5?!hzf`d@ie2YLV&Z=r({?FsW&s)fY%SttFa&+sxiDH`d=!T8tv
z_Odcf7DRCCWk43mBO+9(%%{y941FC3v)n`Sx`1|_z`Vx0akjT&t~voqfcEbMuR3i5
z%7t&3F7+7ZKavV<S_WA5fu69&raY7C`{z+0Jo!plk!HD9oE|LfPQ=D2%dXUhlfFLX
z&J2&nnc~Us2BtvpM_%{9sHH`xmXF9XNINH=yKcojczb!XpeVO<`)$SwPTLq=J<1ub
z=4+*`gj^4VKENP@Nt918%fC(|0o!AH14Vr8iki2~HDNRSft1$jQ10)A!5l0C@M_sN
zAJLz}VEorpHbyShH4>9v!-PPA$Onu}8tuwyXhzsT^(D`E??)cjS=bC+%7kui(coOX
zG;}DwqRm&A&h9z4d=~SEq_ntr%8z0PHy`gYDg)~kxQEsjS#JH_b+=X+D=y~>kA7+-
z*1d-DflmqueO1bdC+`@;ET4LfK#oI+Bd=lSK*cpeU3QU97ensNuNHa_*-TvZ8P0)x
zy~NLk<7EsyhD%nDXKdsY!u!1IZw)buePl;zO-SMVyfkj&x$JUa8t@I#TQ~9Sb2%{O
zSc07=LJn9Ro@VW-9%nO^0xAgx<~bOL^17aCMj2uGQq#C8sf=f07xru99(al=RZYUN
z7l6>XdkZ7JzVz*{n>95l@)~2~zt_%Q(JB^djUH;7VB}gO@OX-gK)Oy?>~ft3KA5-B
z#>O+AXfib`_VWJaZzrS*3K12PHbx{~j5@Vg$Z<ekO|AAT|Kg=8#e#vB$2+bt`cE|o
zBXxZ2unD%nBLKq&I9A_@^hZlQ(hwIE+&7aO8($hBdi@rIK-5q@NY`|EsU$fOK|$;7
z3lnn^maY|}VjPe(n<mqUc~i+E!Pf2b&DS)aytMejp9qjJG;{f^F}?tiDFY4%B-5v}
zQ5JQX;dF2SHe6L3FhA&BcI6q%)3>XCWQLo01FFp|#LuCK=g$ZeSnJ<3_k^6L2l1ub
zxq+G7lB;f%ppH#5D~<&o<rw6H_zOR-cI(F?A4F?MBLy}9DrCaQVway%;mf1{LL`O>
zm&mPyXuh!^M}=5`^r%0m;bl76-{=e!bZjSNY?9t8KY}yeP^Wycc~Cj`#x!!CzoJOc
zx|(m}=bwvB5A$gy7*DLr-5Uop@wvUc9FU6^>ecXe1;Y|(t`=B5cCJ{^x?H0t4Q>=V
z$_NFeI_L|n+z)ZWCM>n4I+>?eah_RPh^vE_Gh82*X&|oZsN!6*v=LVsUbl##h`fyo
z#3UyqipY___Fw?RLWRv&1fE`L2xRe9_vbd~8&_#KAkI14^z4f~5(?g0tL49=6g|&f
zAtY#BOo+j-Q@zDRV4?@kH(_YZb9^hIXRy<;b&n7z?uh3y59+9aqp51D%txM$(AQ3{
z&j-DkrmAGE%jKv$xe)WUDvs$<XT@0ebS{-3{<*cQ^&a5l_L#HXd<;@rpd4A?4({Bp
zDorKTy8VXAzHl*^Pd=eHa5XG<l|E+592=A;nzoce4RoqqrBT+Z9vkuH&<{T&ZFw|3
z!G?2N-MF?W7|L2sLI1joJ2-jUz1GBy>W_L_3C8<d8sIZFwGf$xRWQ3%qxxr3zFeAK
zAQ%0V;-0_3?UTsjjx?Tot%!DQ!_sEbT67tb`4hjS)b8ikdGGzE=?h~V1Mi0Q^DGCT
zzkL%WBQGb-0%=xtzrrr@g`I3zDgN|+X#Rau683J#&h~Y9&<o9>PsH1ID5Kg;X$Q*=
z9X><9^X6;0h5a8WA8nk(4>Le3Sk8m|`iO-$9_6#UN}tp(M)$HE$^4A%nW#(9k>mfw
zJ-dX&r%VDbn*aDUwUl4;7`Yct>m#vcdGs)!{K4OPL*S3hTak2RNS#-W*Q<<9qX6Q1
zj4j%?(l&ATi=)WN=b#^-^|M#wiwpd2G0l1J6a)vS+f#2?KWF=w@BO=`cHHU2p5*TR
zO21_DgOJWPQkqxz+UrFNT%QeTd;jH{U3)AbG)DX>`!uhtaqd06K4Nq}_4b>r?I8xM
zwVIujX<Tyd;|hM&!kveqZyGe0K9?0^p68&4?rYrx16meR+mdR}Jr47fChdM-Ha^es
zfLyotbKrOQ$&HG{xR+JZtv*v9f5F(N2gF4?1|U4aHS;U~I1V4AeG<&7k`7#l=X-j{
z4*kxd*96W?#hlnX-^WcCm9r2fqywvcd;26Cj}d43Z*KM{6>|S|fq|ka!#XB3(a^$k
z#&O(xxhL~>3`{B(;SXx}{J!Z$I50JF!x0Ou8nXO&3gz$k`fug@(KSQh`oL9_ctpXU
zY&96Ci4~>c;@9GR`StKJ#VI$lNP>9FhsPVw5n<>KpBU~nIKNcfA4C8>`H0gohV@<x
z*tPB4^cGu`^$l%meGmM4xGV*4*PQBiZ<k1-G+f$vJ8Z+=viFHO$U<sKLRzUOE%N<o
z^P7zI-Y@>Z+~AMAve_z>HGR~ocW?QX#bCo8M{KL~h<46OaEe~T4N6pLHPo!5CPx?n
ziA5`?&CB?|*_qECtZaS81Wfa&6aaKDx@5)|<gch&ZVmronhWtS)x64>+_=-9H1n=_
zv)OW+?$CTMmCTw~{(lb8|7U#q|C_7615S^kVmQN%*lL{;{@)qw?x~OXD*}6&v=8-n
O0F9oukygD12KZktofJy|
diff --git a/img/dist_filter_2.png b/img/dist_filter_2.png
index a31388d9d9ce499fc265419d80913c158fa8e6fa..c79243c5c213571b4308fd8bafd0f9dcc9cadfbd 100644
GIT binary patch
literal 5465
zcmcIoiC+_E`hRC8350N{9EpIUU2nmFED!}@yNZY?C@O~>30Aan#si3u$gqMe#iOeg
ztRUE0L@P(M#ZZD_M1>Yqs;nGB$N&Wvxe`GV4Z^;myJi1?-;d9S$@4x}o_U_{<awUS
z+U4tIWMF9k05J02w#6R+gam-HR3C$!+A}5d;m6)xJN9}y>i&Wq9a9{G^Bq&9kP3E8
z3C?d%(Xmoer0V=&$NYS0e!E(l;wXibx?QSPKYjYNJwLcz-QKQNLxH+IAAYFSy3@?e
z%-&SjXV5>Lzb*JU0LG7W|44oOt%CrV-SOV?>0Uwph&(H&r^S40#PZSaOjjQ5dBed@
z&ZrHSP!4?gD}Qw1smbuh@S-<oWG@2)^|xOiO<fko+ytVmh#tgt@&8`rxD-bNr}m&h
zZuk@Ik4p+|`u(r!#<A3oMXn#!Be^Gw;%7cK%4-LmC4G&p>>4^>8xCf)CN5vxXk7Gc
z0jVl{P^}utOItg=8nNv%<kw_RYvVH?sazEoTt{r`$3FI!XU<F=33}tpol=*d9*pcB
zRyW5NKbRVvnVIe#bWU}@Q4pstFZ_@~X)BTt9}TBZZJ8RB*wptm7aIG>o4khOGZ{)T
zzv`i1&7$~;Sb?=iSAR0QX6BuqW82+G!s~MLsM;fSW9recjgE;PYN9n`Jgt4cHqpjo
zD_5L78T)qChg=8#%+JZo+&V&Ylj|BEjV63k6UKJ4&Pr-MVkC}lJfxJU_7r!#R68iu
z4t%w==3MHwb;QgRgXprOEIxl@tE0f}NZ}!Vs^6oSL$vye2?xPc@{@$yOs)GV!)bSP
zvEoWSc|8ZrR4$ozBaM6p*kxWX9@Qo^6doSG@WSNDVEoi*{EX#wB=KLHl&V`pZ{0MN
zCbAu}#+OgW$>!dP_w}U4P2)V&)O1=py%<T%;yV9bPb>i0@FIy7F1ZFQOn9pQuM4>|
zA#6fFho(GJHOk=0&sW!fua8yVD`4_o(8BTMQWa|3c13dDo?xVq)S&N2vzWYV;e)7c
zbF_S8EogFbIEmzZ9>65uJa|E^OdLCTM<1K|*0UM}#x7+OP9aZ2wy;QPixLejO=`*p
zWt~xB_|QLNU)X(Os0;|j*Z%0mB!7$Ta==2bf2{OO_b_eS$|OsIRRQy+x9~P|amPdS
z{BRAHwo}y1NL?(x8=u&H0drPUIr}DMzhod5%xQaah3s~Q2cz;IY{EOLFE5zIJKF7s
z#{DZVP;(m3`+!<=L&%caLaejR<&HhkQ)`2DcDWSms6uqOCvP{@kHDW~cu?A&xm=rN
zBg2DvJ7(Pg=ad`UJ-ban;=s=3wkE?KqT}>zOixvieraAbmZ|Dr8bCXrC&P|TSa7ys
zUgM{Vt~|a+bT>=bp+V*xy|KWlWi1rn^&;(otmSErs)8@fEGWKCA}yNn4j)&Y@*KVS
z3ODst$W}!-kR{dJh}!)`mgq)e^LFA$Qq$)7wHaxCg%&9i>k+jKbT<&ib}dFv2aCrx
zq;^*ky^W<-A`V{*!o%mk%<Rl)L|Z4Qf}@^k@(d)=n2g-_E`dpFbH#J_wzh__cxn*7
zE0V|UzDg8{kW<`PUSRh{;t3>~L?g!e#H;%YqECbrn2_e?l(Mm*)rS0Spy@j>ywZ@r
z1NiAFuL9!+9~OBX6d00Di1p=5_QY<)&76ovIkq+t_#;&Ci8&7vy{IN(J+ZD9=+Tso
zC-6<_$azCDkdC<OW7d`g+aHZ%hnfU6Fv)=*@jM1*ZB3YDy+^06Sn{JI@Ee$B-3Bf`
zh6y4;W@|ttOMIP8^n&mZJD#Vat`^kmDV^}ZK!c!WNvS<SM>Q8$*-QWB^Q{b+WTE43
zf^DW(TEETwvj~yhq(5D_sC8M|hC6Sse#a&x%hFCBe&i&SQ27Ue!k^M_L`sx$7I732
zG8CnVwFPm5lJ>r&VYepE7~bRJjIl{@o=zT3CPLrGSj${d&L?<%w{#_rp)f+8p8_S0
z$$(;|-te0|8<s+Z3@;+ow-HzEO}&6|s%EJZFJ3WBXhw~At>Dx)!#0J>2Jsa(aUl!Q
z)AZ9Upo#;o))OCgqbZgTFq9%-MG+qciXy$IYlC|!B4n|i*b_);N;^;PDH}Hf-X$<;
zL@wRQ#mlqM<1gj&2xny2nh<?21Kf+${?fJ30acs?3bD3T_ybInZ)@-MVwQItsF9ex
z{=8r=t_AlEcnXb(Z$kti!&Za{giyutI=MMfrh0zHM|X|}T(oTs9)@Wa=(H9VzHp$3
z_kk3ea^CEj*nE``vId>-QnbwhFW1?gM}$VaJ798ct;=e$F4w@B!K(p_UrtapS6A@=
z9>1NZ3ml1I5pjsn`=3m5DQ<_g-C-%x30>uonEe}h`?xq8i3xG$3D03|tMR5^KA@d?
z;sl^I>&r6?YGIP@_%k1B$q+-iA&j(F>?NvX@NO$V&B8N8h&z>^4JKDTcbP91vx#C5
z=Wof2<l-hkvW6l4kxldvX|iA(;+|F+23#rj#51%~^beW$@2m(QG2z9MySR9Vo;_iN
zRVLh(ZNM)>x1e)99-=39=0mTgm<I40L0ky*cMj7m8DS__N05AIL{qK~BL!SMkEYy!
zThaKB7HAHEoOjep6EX+sEHjh=7eg3+2#x=Pzh(spct_P-T*Ehp22|zp-Q+SEkQl()
zlD0AxaTOM{nc~~a;O7Aujo$&{b}xpC6s42UN{@dU$cpsye_@g|#BKXL&Sf?+iMSak
zT9BAU@F|H~MAR)GH=?ybbgjel^u$NOg=Xw|zZIhdLM%*om(4HVqcK)Q64EK##@hrR
zw3pcP0fzVpP*BC2(}-l*!VJXOh};cQDB{?`bmYDf8GRAr9Bo^UpKDxU2wid)yrT-X
znRD{&i53Kw$Zt1sPu8A}IGd1rfGbse5)`7afX^_E8R4sPYjeY!kYQ7z9}yb!3IWtQ
z4zv`V7S^^BFKz5=^iF~@Z!^x<Z0OMKJ9_>zCa>f7I3;Fj<1k=LLZW;}=SkOpKp}>D
zC-6Ox2=L|N&rprK6ZxGCSY3XVo8-)*DmNb>j))NF)#Yqr1v2%O8UFwmw?j1>@M~z>
z61?!@TPma<%+hGvTHFZLtp4M8hsCqw;xKwU9p3?$qix?lt_2|!<>n|N3nqq6N`Dg~
zGCvbexjlRU-F)Y9;IGhDSON(5`)h97!zV=Ovw#DMLx^yvDF=3$M8s{GLB2nW_Xx<o
zGsuVCrV)%@qWZ3~M$d}50D=eN`9v+S($m2yrWSOxEFY(~-H)jTc{y<(7K8h?u#uiA
znxM9Qe^xrM2XuV(6_a=4CCq?-I2YfEYL+^Xk6{MBYG;xkizcapRNzre$?ss250<N`
zg2U}1q$@C)i>s;1p}qLH_SAZQ>&T+5h~fgx{rqRlV^oEoY*qb};*JtS<%xxO{a@t2
z-p^ZUOlGFZ)jeBVt(`A@3YspA-=6HKbH<-&50V$=(S9Ent1r-|DX?{F`8-i~6V1KG
zzE510CfcaXt#TlW)6XHYy>B!6sfb(In|QNBsAk_^8$Ko4GinYN{0VdW`iwns38^kd
z3CmhtdpQph4al#}(vbo>xo9Z_%yFc86D~pp_dmcfgQo0{0<BD%z#Vr9+{_}MA;SxZ
zain?!9){|8dSV|?bGyo(h(`s{KuS^Ofi6p89Wu27m(ZZBDNxQN-#2Jcw-6B*A6THI
z3RY0Lc=07jtT%$h0u>eZFD}kq*~?IFSr0>cJaBNtZiBdC8BMT}9)TbBZ`+xPFqI$u
zuRMnC!daxtJ|;~Q_9Wv5-lz$Hq^KwZfh{%yXHt}Jw!hD@pb2&-KvqC7WCiTw&?*b!
zA(=~4ez68+nj0hQ1{un?QE00fO>h#hOK!0VpF0Mj{w&fQslIhjwjA1;5<Y7TLwB*r
zeaMuoK?c~j{><MLfwxkX(kN|<DGj#C8ahu=opwi?15dG_DZr-kial`$6Rg9pVS)tE
zL=_)@PM|U!kp#+(VUF==I#%tD7h`ZZXw_4m04aLP51_!9HwMCOh-O_Y8^Qz|VIL8Q
zU}>OP{H=@jgcT}Suj6fIq)80}eIb$ES=C%sXNpySLyeM)Z@zd86;pl~AE_E0jd$t1
z9*+q=3v+qWxvBKvvb${Y`LUwW*9qp_vd^p?h^<DyHm7#Q{qVfcz;;;cj=##DKS`O6
z;o^Oa`H7V26OhQBH;L6TdGFD5Ij;JM26?1LuzA8Jq@+M9Hno@Kf`BGC)(rePP4%G9
z)(}RSU95k0Y?W@CaeZ>*bD`Jl#zRMYU5$CF30(Dnp_`)pj9CgTXz?98YmARw#Ol=U
zv+Iu&PiTV%ZMs|R4LoRK!G1#HP8BEnuaqNG)<n;0q*uF!8-dGpi_)v(=5h41V?)1B
z<N-~4SR}p$S6zFmdMyB%69I9Re`}13z)xSzk=AT%3Wr0@)yk;Xy_a*}qP^Oi1M-Q+
z#;6f98b_QWy?y8Ah$)S8FHio|Q#DD)tY1gEY`USP@<*zYs=3<qJfv4DmQ-`4>Bzbl
zi4l18kHIX`vIwhdtt_ntfk$rG6Opf181k<<)3MN7cD3hlRqRVz1nyG-@wva$f>s#-
z?nnYEjqGJ3`{6|QQw!89-&YHmG3WTEJfW53uPpMSU25|J!ZT#MA-_F`E&+9|#h#Z~
zWP{xjT4iE<2C}Y8>fH%gTt$BtV)3~zi<fa|g#%%^b3q5zYt@`?Ppm;oy5Q^<c?_*{
zS)<G5gWOs$^c#9MK2*2FB$uZn77YkX@v|;t@51%55;LO5Nw3bu8Gn^xK?|C-sjee5
zTmQTEfb3QamO}Z*&hxGy-6&19bKV~dZ1`rL(2??=B?)T7qMFSuxK3D|i`VS_1fp9<
z^KJ1xv#PCzEQR%)7E%%>^{c`-p3ojj%qp~h;s34~-Kwne28Z=|s@F(w-Zy7*bE%xo
zBe-hkeB$+k$qG>C;t8T%x}(9__%k<!NXzFhlRWHY^sAH9pl=o`<b-C0K5wL7s78CO
z6eVn;FHcRaOpX^O!SiD1VVqhih)q+S8!crM6<Cl|EtBG^bH=<83k@2&0j~(@k?OG&
z<m}yB!W?&g!o`OmC}EtItnOL<NX~mS6#g7Xo1Eo;4t!eeCa9H%d`=?k+It2W{9gi;
z#{wcBwCsR(73Z1cfzo^#U`9K!iOc&vL`c~lNfwhdNRS7F;mwkI3vzFQ$VAbW!QkIq
zmm{x`-v@^qel{gbbg6n>@s`T*hG~4tBxiH?x?-HFOPQWH(MW_mp9eK2s2tf%8Jx|<
ziLoN&eA!(#@vvgDPoD7*ruaA)-yA*0;2$)W0WX?3aWPcv>tXO`p^@k=6S9s3{az}a
zOJAr&{cl;L2l^QNU30=XXgXJOsC`NXGG`fqC`&Oi8}0_U@6Ff2B%IC@O6KqcDRW#U
zpB2r%2MI@bLiL=tTsXvs&0(WRuP%xE$2SYw(^T?>bKpA$O=s%?z5WD6i{?a8P&D_g
zxeHm`)Biy++reIDG4~BH!f>9IqD-qSG>s(x6nFTT*X2D_8EM(MM@7M<GHrZ}bnU|i
zjB6*2&Kq0nb^o-T@znP5zH6_Rj>e=~H>2#}#-eJUG4|$`%#6!vt>+$`;n(;kx?1~j
zKO>!*liD|}_<8E*8e3&lmpW;oLQr&;7Gf^wsJz6G2;u_G1ViFIHak4ZzWOD|!;mw+
z>{R&5?7KIqalgT>`PIgSXDgN|WuG24JN5o;$HShbMV~0U90rn?6bOzcJWSM7J>a;G
zCnd&}>|2gJx6c{~=M-8-tK0te?&v!w^gio&=MlgN9;^Q_>U?Q!?~s&M`LXG@se;nr
zoz6ck5D$KdeHoT^O*D=a&J<C-)|{zm?=P<(9g(7(gN425i2m?sG#Fmrf3x-!F)iDN
z=NVwaH6N~j(;fHPh(jc~h<mSu&|UUt40LbRj8YR99sg4V?)Sux9Q@P_F;>ax!<+h7
z)Pln8Z%fLad}Bdm+L{WIm(kM-x9&IL9D3Ye>q|I<M4Rw`{kus^c{rLMg(~U}_8Ica
z=u=-lOnDh{u4G&j*j2HRlU1I^mpc8Fa{9y%sgW{IB5S2GLc(u0|7p{i-)O#X?{y5V
z$85^#Zl`hc=jZAR##Ux>48_?U;fypBle({kNe0?{@h+gT-?(&L?%2BSm^%mLYsfu^
zdryQ!O8Hg$9O9gmOOK^G$`5s?4;^@IL7qk2{$ZdDh)CcrBG1)=TsoOM2;xm?mA)Z(
zpXf3!(Z@o+uc#gBdKdPCA2nQCFn!M(VGGro;<kcj)ZHoN`<D70ozcmJ$j>JAhYzO%
ww)j(T`pjYY|C%btz1RQmo5Q(xh+2=qgn=i;-e*nVyG-Ei>AR)uue`tg8|_BMG5`Po
literal 14430
zcmcJ0cQjm4_oy<X%m_x85TZmcLG<25FVV}08g)ePj9!!IqLUEOdncEKh+u>u2%|(V
zBhl-deBWE^t>0Vk-}i@gcc1N^bN1PHpM4Vabky!Z?m}>IaPDZRD;wb8;G%JGKo5xU
zv6TMUY8Va<o}`|Zk&3Xeu!xAr<HwIhMMcHL#KgtLB_t#yB_*Y$rDbGfWMyUL<mBY#
z<rNeZl$4Z|m6cUhRn^qg)Ya8BG&D3dHMO*~^z`)f_4N%53=9nojf{+pjg3uAOiWEp
z&CJZq&CM+=EG#W8t*or9t*vcrY~XOXt*z~|XV2{H?CkCB9UL4S9UYyVoSdDVpFe-@
z;^N}!>gwj^=I-w9;o;%w>4`ufyu7?#ym;a5?d{{^<Lm3|=jZ3|?;j8l5EvL36ciL3
z92^o75*ivB78VvB9v%@95g8eYL?WZ2qN1atV`5@rV`Jmu;^O1u6A}^<6BCn?l9H2?
zQ&Lh=Q&V5QeEI6ttJkkzr=_K(r>AFRWMpP$W@Tk%XJ_Z+<mBe&zIpQ|FE1}YKfj=$
zps=v8sHmv8xVWUGq_niOtgP(q+qduDy(=#-uc)Ya|NebtW#xwtAF8UVs;jGOYHDh0
zYwPOj>g($p8X6iK8=IP%nwy(jT3T9LTie>&+S}VZIyyQ#JG;8Ny1To3dU|?$d;9wO
zK7RZN06>3#|G>b&r%#_gfBrl;I5;#k^ySN!;o;%0U%!rwjC}j{ZFF>WY;0_Nd>n;B
zO-xLD|Neb)a&l^F>c@{C)6>&4Gc&WZvp;|SoSU1QpPye?Sornp*W%*h($W$djb2_}
zURhaLU0q#UTl@X{_xk$!#>U3x=H}Mc*7o-H&d$#6?(W{+-v0jn!NCCrgE>4r{PX9}
z-@kv4j*gCxk55icPESv7Z*MuaU^dvvM2t{3^Txp;9r*W$JLLJ^9tVdRM?+b`C?J2Y
z;8Mv*<4yVA&w|{-Ut77q_6mOf{8dYlm?w{?krok=7(oR=GADvrxVe!K2=aYAi$AB}
zUn8@ccfM_%ydwrw)YLz3-Gw`p7IoGW1Qz4J1cZHft1Yw=`sQ0Y2TQN#$n7NOD9sU$
zNV7ALSO(@vrKuJ-uV48eNHt|Q7}sN6ANfCcb?B@Zb~g*hQ@5Xgs-KNy^?d?m8=bgK
z;vZ0&n1(hkF}VO!XE5o!t}lHa9gY?h3*J0hz#>O&{e^v>+!V6%kTsx(zYlqPBNs>B
zUHF^BGbEMfTM8U$!`q{Bukdu|w)5R))eM=YwA>Zn?bNqh%h02tlcAU6CVX;s$C0NA
z;q-xrSL;RDh23Wi2V}SVoBqOOw@&9`E`YS)gR`Q`xZ7j0%O%E^wvgEkSjg%0Nw{gK
z$a$m7<<#yUVPmF%!n2pUzrYTCseCelhZ2iRep@Q%V^U4w)>kDo-(n(Y&3vnnJ)ZBw
zL&84&mGhN8ChEP7Yf<BZj`3OCEFY$!TL=p|JdQ2jx(~~?9C_@9rk=*#$`-XbjD&|<
zR4$dC!!Rs6a-u}3R`Iu!MK?z6m<_(Zf`!JD_kNDQmagXT$0=_`E`1NfEzaAIWrr>Y
z`cH3%|1#Ykk(mq1X?3dv`>qLu`(N}~gjLO9m3~K_F7bsE)X89E4_6~QVuB?`kX2b>
z7QZ|%SdzpZf-88hM*fiXG#WGg;Juj;m6<<6e_k<J4A;7sduwwKbP#yyUwC^q(&q|X
z338p;tas0ROIg*#io07N0{Bz<tbzvrx%9X9b~TIdwiUj_d{*{ZcWZ;N86K7cTR2DT
zEER-UIC<Xab?~|#Ppl<DHZTG_CKEhl-7|)sgpA;_@teTIVIE}Lk&J-my?{eGn(+5v
zZy*`Hrq*=(o9qXfwH&WSSkhwS#b(i5c;NSa)3Cv9>|AN;4LWw)EPKaM2f4X4lpE9e
z%zLc|T}N8G=`rw`dNsp8Pavt)pvM1@Y4+}7eS-g+`aeYe{~)Q!CY!i9D$vyEaYp*h
z2Uv)d$zS0Hu}u<0^>6|?@YzaF(4NkB2DK`fK*Mm~W^JA1?tOwJ6zd}_N_y(=uq2ul
zpQlDP$Y5R{*t?i11FzP*0tbn&qmZM2;7*Qx4TGP9jv~I5fb&*u6INpz?q@T~uz5QJ
z4+ErL75PriMqd*LnU<rFN{_-T{;d81MuYCGnsNl$V@Xke>up_ai|VY%%D|fqxr$KI
z{&5cX^S!XWAdP^e`$hE4X!#Jj7mh{#;3-$S%fSgHktmwoBL82N@9Y$IgW3&J?!TpP
zhKGM^4Y7PoZCOXiWt`Crw<#3;G}WcF{Wv+F3B~$2lBPSCsmfqpYxWqmka)j}xEY-i
z(lX@rBj0);=c~fYX^=TaQkG<GZHbi6FY~S0QM^(axDK;CVs3G+Oc!*g%n?Us6Xgsf
zeOszb8TAzNc4#TgbJ+BXQ-xZ0&ofxe-78;Ec_vppbU*Dq?UiG(c==Hfo1e_?)m@pZ
zSITYZ+}^bI4+#D$(Fs31nW$IF4QQudfm|&Eujh{u98j5rSISlBz83MeAG+p^JD*P&
zWM1njze9^@)Bog~%-RbQ_hOLA{zq9EMD&q^*?vC=5_%A?3RIjleRrjcYW3gT0&zgV
zFC7~Hy3!qMZ=QoVE<kdL*ZJ?@tb4LSstO5`K$9qRDfs$PKL1gkPc!jN{i?J@U#!<I
zXOIC9FA5a>Xgt{oOuvE1R52^(qVocGukW%x$MY;s;Rnc8a~Bq8wVFN>PLw{uXnkn0
z7{}G<ODZ&%FV(;VtI?*|*s(c~%5=ypr=!2xDTm!zHDU~U5gE?{Sbeq+yTht9CAPH*
zyShk+6Q)!PxlkTyO9p}3#K<p>Ej3I^I5vwmVQ#GFvCXPxd_l2c33q|jVCU<*s-v<n
z9v+nxboT=nK=pe6bkppDf)5!!#&+sTm(sCf^;{(h{YHlrnKIeoiK6_al|dB7I6hEk
zo9`d*cc1SG$VU9DTKxq7oa!7WEMMn}@p;O~^S&~UuE<Stty?TsRrzva(n{BdXdQ#a
z@+##@H@Z^alP|2UZO+&Q?D%ie*TKNZqIKHEW$szJ)Q-5qBkpbJe$v)@)y#l%eq`YZ
z<O4M9jg4x@6T0}r_MOdKfwFdVRf7pmkc`%zZ=x5k_H#fN3h4;FJ+}R*e~icY5jlsv
ze8xX;6eMx$1r5X1w*pXdOYD|UMbAEGH6kxt4fbB&iJM^jtlJ>L$$j>-R%TOf$$@{c
zFsfI#iYtJ#m6Feo+COwoL=Zv^9;ezPwHnQ1Ei?PbjRS$)Ki|YN-*R+;+6D(^P+m3i
z&Ig_ItT)f<&zOH8zJD2iP8{~|s5~r+SeXx~Bl{g~a+u4+dN}Y(J%eJmHFvi)0{(R|
z_KWf};H8ac$W`Z)Pr6Hw7gT;kcz*f=DZE^&Jh>{5EHb&@WoUtdDCrZ5tMJyL_XlDh
z{d`(flx8*kV&jKp4WDwE5KMieZ&Jw`pyxRl+P^oXY543-$-0(M{@As)xj7KdCqJJU
z-6`Ddoo0+^xUKCiP5%_ESyPQH;wXEnqS}}4T}ERD7AsP$N_>?#5)zeWgb5L9db5}B
z72bbZ+x!+u$c-vxDX)0tYLEu;`y=i4Dd$&Pz(e*5G!uifCQVM_a-h1U4nU|pt4I5Y
zBcN7No$KsQ1(HNr%fb8GlP3V<v9sE=dcksbW(M+{+P=7?H_pab2;#S=dav<|DQ;iA
zJ(XnurTZ7|ggT}1G#CZsMuvKCNazRt!uK>i#Hv{{(+J7#a#mIdaw$Anym357+`~ch
zH|5o&Uo#^?hW25irYs=0D?e{5FVJ>@F&B*04XC+v1?V7wkKbkLVonYATj#aZ@GIJz
z;UxT0{~qKLa)$|;P=bl?{T%#!bf212#vc$8gL%BD-RYjrw10XBE<+!2RD%a*!=uzJ
z{}A7JaLkdgJ~RxtB6bDDPY!39Yrv~HS{n!@b0~p<s5(=22(s^2%t?PTG3hB!3!Lp8
z>zdbdKz(rQHu^W^?)q$bYJRELcKaG-i0O8eQy$&uyHu>xqGsbKm+ZNn!bv{JKI7jy
zzbSl8;{T@UAe5x$hz$q-B>aQl7#yHtU#o{K*oXphYo#fI>*R-x^=?RAh$T#tMb!^Q
z@GuYFg-4h!ns$~bA~+0^%j)P(4R^WFpiI|7Bvn(B;0t#ri98wxIX$s17+^YlE0eiK
zaW$3Ns_+r_`|8aL;$8%UdsIXu%hxY!6uf>Gnd%9_nsm8FNT=^8e)abZEZ?-a0jo5V
zb|}ts4Vp@jXLhl-4oK@ZMo>5xG+ti(6jh(}$FErnS@)yl=_^EQvXDqFh?b-byS?Q<
zF`v<60X)-C{G5miRF>Uqbxe&*GYN#-;<;hEI;M8snI2?0#x7gL_7c71jzzS>lz*vc
z+y}J#+Xz!5czX^*W6n)~pS++O^two$_)e77K4T_N@O|m^#={hJzJ4?5di#nvl~TG2
zMt9qNP>9s$1(6O7Kqye+E9exj$;hq&8vk7CTR(Kj{&I~Cwh&5J6D3g)+`-0?fF}8U
zvfaKCzLA#CfIO(wkMNOc**H_?h(qhvm4?<kio_V&F>-<W^gK%hUz*rp!J?-gQa+x<
zb-QoL-MXS^`sw^v>be`#6)^T(=0vV{b-bXu1b~#0#a2=Cgn&>rb0nJMYt7NY-SzB3
z5b|YfX^h}H_YeV^f+*~^;hC>*@|qgMDy0n~oZXzrPDkWVvTg{V^O=-Q`(0Y$=29{S
z8IHh?c6s!Xd>>gqw02who~>qIx@TA^G>r14Xn)Sy!5IN(Cwh$4nbl1~vrpfTwG^sz
z$xsu`HQDf~MJ<%v5+maqw8?8QSi$;R5=3kHJ<hm4W8ROnre2h#6`kwu%<iU9Ap_kd
zYXBEXBw%^ar&p|0<`}{}1Hv)OhABqzE9PdjSBvol$Cu#EGX-ZLFtO%n$b0L&2lg5)
zXI%5bc6sJXiOLLdN!$z@%y(I3D*b}|J2at6Cp_lOGbDruE!0(p+UhWzUNW#ijX#WT
zQb_gwE-d|syvJV}D!5@UbxH-MaWmPIj<QXj@H0J(Ye(8}w=vEs@t)FsrVgp1!OqKW
zrhp;NTJ)cL&i6nfRx6BU?Z~GgaiX#67Mo{Eroc<19oDW-9Npej#`P~wxw609<a(#}
zL&i+%d`hwhJ;o#wxdUWzYml)D9xeB>U{C8UYFcg{3yBkQeSm`x4EdMOVmx1!LdNv#
z^)W2RJt0w4PZ2ccAZq-#?)4#9@gP@#XZD<c>phz6E5y&~P}@Kc=;OJh33X*!Vgpl$
z^wIGl6KTfo9e}8%KiJ=s5}+>n`%fl2lSi&ok;b>GNt=lKOqmdC@-RQ0$_U{gXAXjO
z3zVu*=hqqkj(zNr<*kF2|3E!t{+r}rk_QPfK9MCnNOI=O5hj%#^_eCkTws+lCL
zGQP_}_AN!yD1c6roE5;jT(i6ZuzlJIB;>ySu@3uxiP>e^<oaV(Ow-)YAU9P(YEju=
z)<R)?@6n#rJo+6^@J=~DQ(Yy!*)S#%+i*ccOCf~~WlvEk8$jmCBk(vo7Dil1{}Za-
zuK5bVS}%ndAK9l})1R`nzzZ>pV-3Y%@6VZ|avBQYst&%3ZG{^R2={EHpF)4b{TmGz
zfPrg$sZqfQ(D`BhmA^5?)#58Vh}u#xGD|_>Av0D<%1HWF2A7i->k?Drgb?SZGoCV6
zIU2vfnVyQF`%b)pI^g+if?xg1Z!6Gf(kr9@&$OXmZom!r8cgk0q}){;ahqi(zyj$l
zdg-Ttcd&9tGoG`3mwOIOTya)+B0$6yAWDoK?;d~h=Mc`bt4Ga{yWbuE!L#KnR=oI)
zbA8OEa1+=Z=+0!sj~g%yOuMLDRa|l@$fmOlk~8!fs@4yuPOCs5^K?tJ$#FL=HdRCb
zCbuX2$M@pPos!}o1_Utu8^RssI_5X%1>0s~7MIIMk&uUHv?{s<^r~N>TBS(h+O7y1
zg-?t(glgI3nwT!rrveRQjD#xOS?HY)G&YYQxE7_fFipg?ICf$bG5dK3z@A{lB%H~N
z5C_g}iF(LW8`Y%Zl`oHt%2|lR7+;>q$L>jR8kT@;+~k9c`}c7ES@KaaxwuodBdqsU
zk@)5X?HoRR@jFEUJYxPWrC{c=BdFF}C<}3+zVEvwrH8lMl05OOXIjGM_>4K%en~{E
zXHlau)AEG-EanWrYEda4>QO<Gx(~4}XoIZyG0Llj;Poa&DN$IdTM;$BFZXj|eUMml
z2_N|jIOx@;5b|FT`@nN;?H4jGH$e@8N$bUA$!q+vxJjsGjz=rWI1^_eJpEfq6+jyw
z=5(e<Hg#?yYgyvq4*nxS&v)0DHo2b>6`C|h-%^;pY$xoDr9+16HSK;CqF?xH8ApdC
zbVX4Rb&U08k4b(a4=Ja1k9_sU;=imBk1{^;G!m}fMkF0_53)4$g7t%P-Gf-Pcr#d$
zf7<({w~QiDSjiuQ?-i5Xs|yk2$UdVMtNQFS|NmmaR5k7!4ixQq14-n_oz987o8qBa
z{%`KV6cL4KGR9f6H+K$1#8{%?G1+&^OHR3HKap4eF;FtRw;?FTibSggI^Xeei&`8s
zCs)GMH)cvCM=Fk)T`(ijSxwGl_8BSOW8Vp3paa9^h<e+0yA27attcX7Q-(~YPi>@k
zMsGZ`>1f!#wGesFm?%pGy28aT>O?TLa-|O!7xa>+YwMc1I^g?VJR%Q$zrHb!h}5M#
zm^@;O89U=lxX%f;XmLS2!5orIPqF(|0KjC_LwOx2Im3V-L3`yaH3xzb)*(h8;RFEq
zEzXo8?RJrIsxu#<9^3L&#mh#*Gj#ZowB6n$*ySO9q_|8G03<FBZBr~ueMjOTmpqlK
zih&!}aVmU#+AF=$yG{5~0kL$TcC_oUGc6lcXW)Mj)+Z;=bF@}-Z3WRt?+TO<=lk9N
zBAJ$Z{zKgxSOyAdQ&H-mhnnht)Q>I(FY&326r<C#8CXe3%@G@3%_MmI_t$dZL?~oc
zU5PvSs}r$Q;#eHM*-CWQ&%7=P0y;=DoLZptZ{Hdnobw~niE5!vn7|EKhMHn9xXq*|
z8Dy9d0TDQGY~{@Ot-~&=NX?wgwBE2MEaUhiDMQ#tNTCVlF3-*5wzF>$)K~T7*gYS=
z-f{m1fb<67?L|qbAqwoTk<^x@)|o$(m)hSsL8{SO|5AwwhjdYVYU%XI6CnRSJWsJt
z9?WBE?R3johta~Lwp>cU(m$DW_X`sn)d#gnloxb23PTO*k?bVn1?Y@|=dX}V@7p$d
zHu)=dwW6pNyeP0g=B3Cd5}*L#r&qj$k2>R4Y<`QP%Kv9*l`+3mm`o3&B{@gv4e4E`
zw4%cS^91CrGB+m`*hC=Zc9bvSI^pz1oBcSPrakivv9J-2PQj1F#u(AK4W+XO*}MHX
z9A~b_9^Q0=KMgG#q4@b|zyDwvy|g_~8eX(XbhC}WARd<%2Ox#dF?xJcwWr`>I5k;K
zKzm2FT7WGMrL<9$pMny}OH4n`!?XCHIAa>vo(H=YVww>fHqFN{>f_R!z>~#w49i?O
zV+Q(io6+1%wO-%}C$Kl_3^SUFKUN`TH|_bwz_Y-)L0%N7hZLT;{{;a>?Yl8@0xnta
zk5;}Tdl=eNX%u2;y1$q?<SDuWN$=;hh6MN55b^)4pC(SQyhkBv11lE`e>yay*{Q1Q
znxbjCqf7VFn$gr5*DTTnBdoHMA3-L^*nnID7MS=RE2@B%_%tes4Zh~vfuEf2=Aiqs
zinUSbR9MA>INBUI*wvoTV1A5M=I=+4!21p-Cl-~QEK^N1sOns$-8n}yU>zVHGxoCh
zEl=~G_ZV=i-FyeHao>BGp9Fy%AwY@FPL}DO)h+AQvm1Y><e!Ma*Fi|n5d^<^`9g!A
z7mi$F|CeJ*sx})mLFKWTd0D8S)A-~2Kn$@KCjN!YcxT(+FMO78<amJs)Ha#tgzM!(
zKlq_(N6N%ImR<#X@CIjVnbJ2g=N^tE>RR3Ckf=f8o8sALX~mu>aX>fUnO$p(#<gfn
zQV)5st0ZNcutN#lovVqY2<af^i(wB2?WlK<W}wIU%iamTRht!a&PNy6IrETcz)tSm
z<uI-E)JR~!OQOnvzZ)xj4#;+EDQVg!(CiM_WJ=!FfG<)0AgmShdNL0F;&=7R(36rQ
zQFDl9Gn|RRqnRqp8Ia{X=F@<l*z$w-qPfg5j=l4de*IK@0m6yCYbd1H<Q($9=u;Q>
zN-bI+xh&dbv|Eehh-lG6UEXJc$V_uy5fYPa!arYFz$|A`E`;NW3sbLENR%<FZrEh;
zDBD1d<`Fqy_}iJaT7$gG?7Jj0QtD+{Onvz>g|FjlJecK73Qhl;O0Dg0SOenW6nxPV
z9b}bCN#ybW*6NARQ^Po#2jqYY9}Pe3GE1`MO9YYeuKhC0kL#&Q6#``h%^U%)9rC`-
z1k&vFq1g5;iz3ufW&Y8Bff$~r^-m@=<v~2HIn7#D_k2BG<G^-}>Pag%bjA1SeGxDm
zShrQp@eQ>k_xq?MoOra=sxz&~?14T9tt>gfuH;NDLgJ9v)`ZUzP=x%vNc%H5O0u&)
zlE}Et)ApO|_HI9zrTv6^4n)23BP5Qf)}%7_yJ}i^IjMjaGcr#Sb^lAdxjLhDjy$MH
z7Q45r0aU(0%eb(Ga`Tqc#|xj;-cLfw;?b#w*nD6?i>msZQx(x>kf}a1VSgJb*@cKC
z>NS2Ts?cxhI3VroV~Y=~b__lhoi~{RRB)6rD1~N{6Rm?q9{7^*EA$~ZHnLf!GM4D@
z2{9voeAqbn9dzNnjbZ_@oZZ}SCLh!qO7x3+K^>1Sn)*;`PiKbgl@o}-CPHpAze-fM
zKd}(9oYk1e><8tqgkD>LSppO)?=spkCAH_h0Q1dq?~Bj|TGODAH3kpX!cLzGBB@<0
zXCKT#ZXY(oT}e3l*mOrIn1K)ZR$lIuEuupZI-u(z%8^cgO1m&qMG$$)^*w#nM<US3
z7$B<MQ>n7;=bOLx=Y2I2w%`NX8&vYN8_#%WBS?K2FU>Uq2elPAunj|wW8%E`G{7wz
z1gux!dCzo02gYQy2Q2aF$-&>`wd+&-(coRL+~Ak~alz#+%YuPABp-=-*_C$`FXti(
zwQWAkZ>g&<nuT`Ge7ijPl%hG*1hRf+iq9IL7%wB$nmfAs=F;(O0oY58&dBfX@1brf
zdR*{)Z+nl)de>NXdd6$J>Fdhw+9&0Truq+u-Cs0GT;ZUlYbw5(k;rBF)!^f!ic9~Y
zxC_a7UD0XdjNcAY`D4GHYVTbek`)q}Hwu!|0ZY`Szk{|uuoiLNPeyYnP96D64Ra(C
zvYutqF)pUF1$U)@lF?+(rt<i%(?ZG-f=DnI=K{7&SMFSDj!~Aq)|!X_K^s??bma_1
zCiuh;zA{`}Hef5B6Wh`K7A{f*#A9u(dPA>X2U+;Fy&{zRi9^9_^k@LM8*`hgfQ&Ic
zv>AeHKeHXlJS5b8XZip__zI5OMN3*BW&SijGbd+@x7wvwI^@9Qf+_%qLfRnegg*Lk
zhgQevK%-lc37=3N*~hPFK{6bDR};nzj0;cPD6U>*$r1+1Bf%Xe^ckdFCzdhTlCst}
zDN+T!C<<OB&+2P@24pb-yJ_tG307`i(Grm6L@RI-twiyu7$jyRKQ9dAI+ALTM^S{l
zgSn4``11lnU-g68&;DLJ2Rm{Wu9JiLIDsWi{oQ>>;y6xc1>(Y%lHa*3z=ltfTQNw}
zt71SfHE|mGzzvRcXrHvDn>;}Y!|w5ki_fdnl~O`-(Ea+&QifB%)?fq=1c5Y>_tHNA
zRuV;Qcy2BImAkS+eBYKFkCIp8;H~Rp>3(L=!?WoJScT*-VO97K1KJ7p#;Eq){wp&y
z>OAl&bca!^K3KrKK%yB<)M0`&c1mUY5Bmy1izQO9%YDP6yt0$U!n<j88~{jeOrc76
zh;-RpHjYrMp2X>%v}H@r`fCD4LEuI8*^<t6iW*(?M^K-p>WDC)o(y8|i^8Ym9mq-@
zp1rgzEe3Ij<JJhUO=|eJvneFfFp&G3e|#}OQL<EuzdY74$WHhSTe07DS*E~$EDj(K
zI_Kv<v4dYU!*4@ci!_OjoPjWjE~^_mCrvCwpV$YmgMtu9M&|?tFY2sHy%Kz`*)3H~
zlCei!yKN=3xHn)Z|Lo?k?R5$epCHh*A|R?Yr}IZ0Nt1(ijr%v;wcKgFX(C#PWkK^I
z`Z>2g3{ez$2D+`k?lJE~(M5*R-RNQvnMvu7+_wVS`&Ry(fWx)3x8qjp_=_)5B(Kot
z;>}`zeZ(F1H*hF+`Bhw@QPWqaM+thk6e0EZ-1!)Es78XjaoNPJj^<78Rn`^1j3!D(
z#~Dm@Sg0Mt{%FO2B=SfrncgLM;=m`J1j}DmghKNaHCz`JKyoxe4NG$SBogy6O4wp_
z<TN#|*ooedBAznl#H+bRT|G#n0174o4pf#X_KNHb!XT_5-OLgzmFWCTCm8l~09DIk
zHzczH=aeVP!f1YHnX&IK0+U~TOBJQ&@}kJ{#v$ruQKL7IoARA}sDvwz*;i<Oub%t0
zR67>$1}UfBJl&Vea1OcpOB>QhUc>dG^MgiNNkkty-Q8K6CjPaDCt)~XW{9nta%R`M
z5&8plpX3@Tq9dPub>l?y2NV<V4O4h|at=I*=t3v%IICs7?@*v8#a#ev>X!-$4m7re
zZt$Jy^P9g`Y1Qyci)lmu9d%ZssXSoQio;n%LLZ`Hd4;&lvPl<_1<ndt4Z}}6m7t{A
zl<v`o^>J-*i8i|$C-vH2ncZNHl5117+Q&-Qv<#w$|F}))pFYj^ciyV|wxu<phZ>hQ
zF!ek4Izwg$%rH?bSkBa9`AD2hKDg`aL{Rl4_fxKbF(l#d%}d6Nuu7L0ZyM~vtr?Q!
zuY7~V!Wx57WqDqx-vjr@kOGCu^R^L2d@(I(%y4F<JeMcXC_os@6(&zh?kx`8*Hge)
z=W+wCz_Ou?DwY%gm5pf~h+w_;!F_~v%a0mV^lFQ$l7>N+PiSuo&1V`jj!xp(#Cw*C
zC7EphZ$OM^>%SCMcC4Nnl(eW)+R(eeeVp^WkpD?Kf+j>ErpkHX+j8uItDS3w$J*M$
zJLn{pIRX?-B`w$cYr5e~_wsL3y_l#3w00A>%T_V5sWJ2-@S}#TdtxPn)%6g*RDQZo
zWNdUDddVQd-1J3P4L8SxwciUgN0CHWi&*7zyOGzgzfZ}U4h8fgt0d(W)e)y{M)sej
znlwp8Z8{%cW45|tVrFP3A)XgEBMm*RvM7GsT4WJ70`Zyl6!qg|os~MqY@Q;WV!Pt{
zwrrUDUb%y`3Gn0lV+&h#ciN}b3d>w)xoP1a9dn<=bZh*j3c*NA+H?LEOL%byq6UnY
zU69TGeUz`0p)fg$gG1f>uNNTQ6(J6hW%w2TF;+kMO!k{DMoU@}E>b4J1`xy*p;>D-
zBbL{2J+!<juM}4#(baUb8n4|D?Zh79NvS3{LVYFiqMY5JzW57WxPZwBQRy900YJ?H
zwG2`Cx<u4?2f_ge*sX1E>8Nm=mXqa!#JCzz;}OwLy90M1OaRjn6Gn(nEjR8SgDf)T
znNNK`xzvdCuVM@gyUGiy1Vsv+3O)4Bh*A}XxJenczfInO&;qAy9)5#wCi5@jy<s71
zERx1v@B0ABUuXo(ZE^-xW#nZoy!PO%OWNlHp1x0!SLXxtgELc+gO+)%F}uO7*oZi^
z2(9GPNcG>A?r);dYCkviR9F(++aD}5x0?yD)hH<dCpdZ6Wt5AdEt_CDzhYr($fAtJ
zCb(@ov)@-MuZr{|X%Ik~Ytbp}1Q(b@C@JxunGr}DmTWFtArmYSmydkP6}?jZ)yQ4|
zKhrn0VphZtB(tZGP6iXN6F-+I#4f*yG)MU3OBytA3e&~961V_?WhESp+PfE7w(XA0
zG}3v=I!1z|hq7QD-u610g7@%+g&}@RLvkp8PmGR{$nI|cV{ijJG}}&CqW9gTdI|NF
zvVe%&db>dvcks?cAY{lCx5-Cafv5eOu34WmN2+HBuQ~91Qw02WP1Eu4O{5XV^MZ9Y
z8yYfR`wG%)9H(O_Zy6%C_X@IDC5OCz_3apks}k+wlci`d>Z_1ofH9Pk)O^rk{F&3n
zP!F;{$pnNg2eqXcAXJznpFHxyxWRO}Oqb$MbXVC`aq}Ye>KAqCBN9GYr^Cjy&165Q
za<<#lfeAmhR0;XSrJC5O3AOFjcc;A3!-QSK7+yKfj{~K7jGkXKI6aF*ldWcW&6%mH
z`@n?D7t@D04Wp?u>u)ps`HCw2KFc&ZM(SXat)CMOMqf7<c{tk`IlP@YsTFws-80-%
z1HS-maltAso+%~ztLducr*XiKjUvV-zu6caMCO6%<9nOxVf8dK3-Y!v(wEF>$K*ZN
z%Zn-9W?Iq}AJHwfF-&J)u8<MTDag|Q?E>GiN7~G_JF;`s_6v=lzT@CKxq^R|4)*3f
z;%BFnHXi1Cm@BcqIjrqTeCTLmx2(>Ri;mUd$JH4Bp;SV4+HQEuU5Mtj3W{MYs}xY7
zypn}6CK_0B!n85JD108eG*qYw0(_kp^oEuiO9|}la0bSNXC$*M^Hpl>rvBXaqwE{`
z^2qQh_R`$RET-WSS{C3}%BDEgnloG6Py#6XPVuDP>umyC&<^7S+b-#CA$tm%zGo>F
z<gYU;72pgan<_5Tqw=4NCr7&a@<_3iQ~qKh8k<VGHq!bOYw{#T8M7ejLjn<IpfslS
zq5x`zd?4JDo_PvT580Qb3eRipyYZl7KOUEAbfB-OxzD1kpEH+BH5@+H!?b}nai%ux
zwO)4AWKm%Qx4f<#q}|Z9NR^nMUU^)Wn~vw(WBu6-#d_=KXk{DKAJ7Q2lewUnxt+Np
zgDe|)$PRhXoPKnbY+!sFI?4pql*;{3v*|vlL4nz{P6XljQU4^P4BqqoW6@Z2dSt6*
zBkk@q&z&nTv3JMkGk`Uw()h@&EN$GIK%982Kz{tO<d)1w4Fq0LxqhNl`b+e|-wELq
zV}q;v*3bCcHN~O{^d$yW5L@m?%4M0g<LyM#Vx{`+jR8c{PInP(=UqPma}4}((3^>o
z3Vu@L4u4q%eUGI$pir+=C-O08>>aIig+(X+;mm1T1G=(jQ$tm_>mVKKu`91D^hRRx
z3ID7_OrgDq@P5U)hy`$vj`B|SSg8buHxSBv-h6lv{0yj=Lm4J}ya@^Z&_G>Hd)YvO
z4ef<9LlEH81f;hQDg!ukL>eRJxsE_(jL!hD-6pZB)F<Szn&2DdBEw}~@Xoa#A}w_K
z$_!hUVvy;3&+%)cTTm0|Ivx%aAuTd=<^>OxRa6WgS0ab>uwTtbM<QTZu`+r{uNO#0
zMtfo=8oA?h$5#BEKQY*8=X03&<vuxNdd4*1Lr7;#6Xlg9lR4mxc%_4B-zQ4hP?sx!
z_v9(zq@H{y=4VhC*Rjjrd`p?1P?$d`s8hs2DV>e{2bzbPmp!5kF*(Sgj_@nOkNpO+
z`R*EnhuWI#P0K}Ji=omd=>{_g?2*R+9JcwlGIsC@CQkWiLF9ht&U?^K;l3hjLYck2
zk5oo1C9cV-5)?iiQ<yh6;MFK)CNa6$@mzY!rh=lHHY%ogutJHD73l0trOVKT5r;K1
z$~;cVma%~mPp91d1RwHOMtW}ZH7UNj&QZs7&>T&PSPQ)$d&sz@psT&gK3tLhfE5Ui
zoiLmXcp!zdKjo^SDF09#DkD6~f_3g$2|39DGlg`Ws}V#Y(=%8~6dCSaT2X%ZXV`)T
zIDLY`s<m71SH-5FOQQaHMb%LW@-LJh*S4eE7b)U4s|JZ#DeLOk(b0y8st8kBPQ3j0
zb{a!RB84fWi#~YTFXuL1dJiaLY9L2zBC%fGDO&YPCb`r&^TON!*-^PaF5T!~p12tF
z_o~g|=Oks*PD8vr!1z(>6Igt>5_|g=Ns!0gfp&xKrBq+w;F%>>D0zaeNU^^fbh>?s
z3Mfy%61nfdCT>R|Bb6dG>Hi=n6Yl>Y$nWl|p<Jkg8S!EurHs_*VdKZJYRJvwJ2Fz*
zCaL2d&VER;wz3M^W0F}%BvB*hnKo}wkp8&+H!^_4jB4Q>SwW;%#o5n_0Ubo^O)AK7
zJQE#TO?+cz^DL6yH&)MH8<EGaM_lflW>jw_W)29g`{$7g1AXEs%4EO9(9UW>>V@j@
zV>^1Xb3!-mdTkTA0Ru@z?z15V>=e%@&MPV-AptYVyezcM_*<!KCDn>_59*T2NH{`;
ze>4UyEk$A(hBS!Gq7<}b=n6?vf-YPWIe8={k^#s#@~ZHkC9xcll;vlAy!6Q-s{MYa
zuVMEu7vnc$#A|+!IQG#Bes&9h??RS6E5Ps#3$1jGrm|A*QTUrz1!%K-L5UOEYOD=t
zuoM;=RS<5I;3ZzH96x3qzKJo%REi!kl)Vx6&f+C1M7=YKXjj~8ZQ>7j$S5=Hysk*y
zfsPSBVk^rYW|S&ILC0pg78OJ)Nb`K`H86+J%>-4B(BV7Yj@RbckKvd?7ZIVSo!76Q
z_e_K2Cfg*>Ca<#h1&_=ZM62o=1H9+;CGKU}xMJ4_K`*u;<wB&0jk>yv-RpOVTU)M`
zsd;@yJHL}r4~az`4KH|q5ufX1fQ$uI1y0fpcOPtR?1lRpH^ZOPyM5K*+6lg2X~}2G
zY$=s=!X)9<dpZ5}dTIVRHM!9h_ert~@O@$d;kj+Rk;_UgRWOEai_K_XxnJbBXI}s5
zt;dzcvCS_FXjaxTp0R2R{6VFl*OUn~D<>zA&(uM3JO0WAc!nl2*E51<6%eOeV9)k6
zBo?r}oZ45eSb^FD*1uSvKGt@b*MDvQzKNB_({uXpCXt4hM`n!K3fp40ML)mY=G{h)
zq(m)qMo69WJn72<jdPSv>PHk<y$$0dzEPP;YW<dQ*8uyXbhFD=Ul)3Ohw=&-gZ#xx
z|GqA0$XL?Y<2S873nONHwbtG?VGR##U};x2scIffb$#nW+rTng0#&tG*;FuS(KLls
z_YqhiZ}3Kvz`%uTJ-ezQi>xOVDGlg0KWR+eWm`N1N5GUq`W;rP<V?2arvoq1zKfz0
z+vDf288%3npvZfjD0&TTyqZ~RULyYc3Rn)Qc>|9rd458wy2~Ks`R7`=LjBEMx0qT2
zEM1xZ>{`@E)c_mx>n;gcXs?X26`K03#TtG(V39n^g4a{_X0^Sxq<nPI9j(UG6G#Qb
zMV7q+Cf&_PRY=&O)-%@^+iqeKSPc%IKt2*_qfD|=fDY=(YVi8Sd;#Bn@K>>5E*NpS
zMbaeq+uKR*I0NKbqPGt6Ye=S@arBptS>;_{YCbx}J5PTrGBv}ANYe`lYwWNdlB6<N
z`<J%BNhW`OGM_;qVLs_U?jJvR0oc%%+NEW8-fK3Vfpws(jyH4G*bg|K)wnA2x7QLf
zrgwjeSt{i>V4=;IN=el)xAgfD6&i0@zxepxqnQu!HMOkIfI2s<(!1A_*z;8w^rn#t
zt3)OHcDhsb)4$p*)_R773)UX**+*X=tv`(^YJ0H`EHt}5C3b1FT%et#%`dMNFj@qX
z><O<G>iJQ&xb&bRsT3WJXKC;6QeQkU0Dg;2aA_oyAbshTpkJ6xZImtK@qoVVrZSpW
z@|eB5KI%fOk?1noi3h{UAf$IW>cOy9=^1hrJ8~&!$fE5+Cu4FOjNTnd)=G_)whewo
zAcb)=`K@XQ(2@S%=f&L-iUfvFwdAkSPPv=$YpMaj03ixCoPt5-9X-xskH-G@k*Vq)
z(X`cd4b=}i(HlVn1&@_!rOgv`)L$smhFsxEJ{X2sAqsn6I^3B6@3`_N5%HTcxzs13
zkwaA|+~HI-Qg{15RPk^I##g&ou8p0UiW?xjwHfkO%?((+oH$p1Ig892FZ=uYegl{`
zx#U03`H90XFR@2sZ%bBRGnIgKw^talVCI-DV;|{1tj*{Gtm{IGQ0xXi4Ku@NhNd_$
zz6S5$Ig^l^VRA)%Bq7X=XsfbKA1Zl!Aa@gCBo7CgIt}shK{V1mw2Y|`BHs4jypjym
z<ffQH1s`#4=6ZM+zt9LyBsPm^(#P`x@7(d%A~(Tg2b+HL#00P=*{})__e1s@Tn2xb
z71#=N1bl=2_^gF(yI5<Cd7jdiImeJ0I^Nyqyq)g6e~w0%i&+Baj=BUzHPTZ)<z*G&
zK9zEBx-Q|WSNnfYrxOvfjT$X36XQp12|VJH3!vJHDeZ;M7t8j)ZdUfpQmfsGQ<d=h
zJTIQsC2_Bve|>s_?KuuFs7K1Ny`jIc-|{o$+n(QRG=gx}Sc#i`*`MiX{CN)5Fh$;G
zQ3f6DKb(IxJX3hHdADn76LBEEO7G@+zsVziRqAKEu!<>Co8h`QAMvMVi#9xe`Wv^1
z<ic1|Qq*2UL+2l+tEo=0$!PX=J|nrVe_eh9fyW)XP?ZA%DR+ylp)je)3AHCSnd}bB
zuq^Zkw06nCZrknSg&Dc*2Pa}H<860pQ+>HpN3)jqJBb`r6ity)Vln&i9TW;BI}K;H
z4=X1GVR9WF)u+!qn$F%aYQy3aGwMNvRxWhSqpBjvIE!Gjx2pZ>yR`QPFRq7p@6NZi
zC{r2PN3#a-OMH26rLm+e9=(xjNMcY1=UiVbZ;;j4X}xjHlVZ{0rs#)VAxvKBXZxv0
z$9D8bO&hYJFQyMd2PM;OudjmbJU{8j=!zpf?~_Vz{{2ita9Cm+2lsC&m;B&H&n|aK
zzf<Klav&DxY4(NNvL6;Agt-k=Ghn^Mv0i@CDE@fx@*Z@h@t&$99?9`u?FEjmPs(EV
z_lY^5w?Uq^9;()8TF#pV1LI{_D*EH+dke#-e5nJ}!te>rHIkEn=BcL#JPRJWI#22k
zdam42p7?)%SDglz6LebK{gBw$#_r)Q+1N^J!XOz~+ZBE_5*MCFe3i0l^kAq~a+xTs
zfJ(6RtLO;+Iaf=gZ*+r1ddS+ui=S1$H|V`(He~0mJEV9XO>BMnWwI3HyRyS^5OS-1
zUV~FNo$GL8%!J=hzF<SiSyGEfz#8~CtejK00pl!i8f>vFu&<>f6LNRu78(hWwQlQ)
z7cBllq=c|7=4)EDzz?;K8SSd8WAGW2`e06Y_PpF|p<@nx`Z=~$wmhKaV-Br!HaTYr
ziP)-w45M6+fRyk9cN1-IvTIwhH=h~(bNgoA#D}UMK2s1B2w+AT80-<3HjK(Lc*r$M
z2|wc6Au6$16&+p~TXmBY`2D1?G@u5aLu<)^7HSjA>%O;-z2%a3RGZGb|LyW8!?NFE
zaX@8p4$T$+$CS;j*(GrjZ}HUqZ(66Yv<H{x$~?<~ltJ{9NHy-Op>&xshy0~?e)_~7
z73cAX?rC9X%)?OdP%~Qh`oLS}tE;>BLk`p(wyNglG?IYR^%14El#M&7)ZZY5r?;e`
z1BZn(>Km@!&b^Y12W<2QuSa)udI;66IY5r!mzOf<GSoW<VbatyX&Q<3KWsSY-G|n%
z9e=cG{6`@B^n9co9<c*I@Nj5Y?9H76P<0HrWqEnL6}@bM1>`wY(g$t&O0E(c6S5_<
zv1STQ?SJ>iQUq4NvIXYqVYyGYIdUhcTwSAKf^8KU=U)#CyV#UM(>o6x_?cM#BVrl&
zLsy92<`2_>V%N3zAQIYFLcdim(0{4+-1;GC9TXmaOsT~<J{swo{4+f*ZC6URMmDIt
z$uk*P`s-H;RT&!0)(BJiEA&8+1DiZ9SzZVad+HN7uE%w;n#|v=O8kCeM&riXdG=Gb
z=jN$A6Yvl`bGFa!^H~OuWF}k2sT$HsnNt11;kwH514mFUc@8DR>OIZpFN>B4;QYT<
ze*Ri!DSa&?Np8{GD&O>SL$%sBhvF*p;HSpJJn<ZiW_1gzQU(93JMK4Qi)VAce?@C}
z+Y3-q86tJp-Fo8-?b{Ltgwm%k=xHr}3D9NyEOfi1&$iW);<8~UDhG{Yo7zV_yPEhj
z^NByf@ezXtN~@sluC2aSaqqctyWSVk4R3tv9u0Us`#oCU@#a)K${>LMMc+OzDgIUG
z`cf~^b)C8(-|WETes?tY!Fgwi5sL5$&f;4Y>!ymh4Kp2_Wug<z%3N*xpQ>T9vpXm%
zxv&=;%};I_R=TYF3k1Qz0b<q)O%X0Ju3@KwZvdFF2HrW5O?}RqO3)W^QkfUej9^!;
zA@=h!gOCYU4pnLZ>fQn)mbJS!%ADn74UiQsy`X2eDIwHJ7DCeYLXF`p%o<^&1l%!_
zNZM4<n1h+mtD=NdF`&Mdm~g!s+6LMJ9tg3uV7~2e&!jIi>osW*YU({(K*@I>|3Ay?
z-`$_L;E$tN%hkAmrtqKE*5M^`XBx^TCfcz*?W?$m0hn$EMK@-DL&_wNr>P)8Bar$|
zjaAQe**yocg1LgtD>wE07Pp`&itexTYW_jGOf|nO(wxS}nmInRv(3MN?B=TMotCYU
z`BSTaFEO_3Gew1Egpt^;8a2NE`{vUBxW@GV1+f9{QvaU|D*un$CmXl;El+iYQ4giC
R7Y1=~G*onyYq4il{|jHKh8q9?
diff --git a/public/07-basic_statistics.html b/public/07-basic_statistics.html
index 0716ef8..326a724 100644
--- a/public/07-basic_statistics.html
+++ b/public/07-basic_statistics.html
@@ -464,21 +464,23 @@ Moran’s I test
</div>
<p>We will compute the Moran’s statistics using <code>spdep</code><span class="citation" data-cites="spdep">(<a href="references.html#ref-spdep" role="doc-biblioref">R. Bivand et al. 2015</a>)</span> and <code>Dcluster</code><span class="citation" data-cites="DCluster">(<a href="references.html#ref-DCluster" role="doc-biblioref">Gómez-Rubio et al. 2015</a>)</span> packages. <code>spdep</code> package provides a collection of functions to analyze spatial correlations of polygons and works with sp objects. In this example, we use <code>poly2nb()</code> and <code>nb2listw()</code>. These functions respectively detect the neighboring polygons and assign weight corresponding to <span class="math inline">\(1/\#\ of\ neighbors\)</span>. <code>Dcluster</code> package provides a set of functions for the detection of spatial clusters of disease using count data.</p>
<div class="cell" data-nm="true">
-<div class="sourceCode cell-code" id="cb9"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a><span class="fu">library</span>(spdep) <span class="co"># Functions for creating spatial weight, spatial analysis</span></span>
-<span id="cb9-2"><a href="#cb9-2" aria-hidden="true" tabindex="-1"></a><span class="fu">library</span>(DCluster) <span class="co"># Package with functions for spatial cluster analysis</span></span>
-<span id="cb9-3"><a href="#cb9-3" aria-hidden="true" tabindex="-1"></a></span>
-<span id="cb9-4"><a href="#cb9-4" aria-hidden="true" tabindex="-1"></a>queen_nb <span class="ot"><-</span> <span class="fu">poly2nb</span>(district) <span class="co"># Neighbors according to queen case</span></span>
-<span id="cb9-5"><a href="#cb9-5" aria-hidden="true" tabindex="-1"></a>q_listw <span class="ot"><-</span> <span class="fu">nb2listw</span>(queen_nb, <span class="at">style =</span> <span class="st">'W'</span>) <span class="co"># row-standardized weights</span></span>
-<span id="cb9-6"><a href="#cb9-6" aria-hidden="true" tabindex="-1"></a></span>
-<span id="cb9-7"><a href="#cb9-7" aria-hidden="true" tabindex="-1"></a><span class="co"># Moran's I test</span></span>
-<span id="cb9-8"><a href="#cb9-8" aria-hidden="true" tabindex="-1"></a>m_test <span class="ot"><-</span> <span class="fu">moranI.test</span>(cases <span class="sc">~</span> <span class="fu">offset</span>(<span class="fu">log</span>(expected)), </span>
-<span id="cb9-9"><a href="#cb9-9" aria-hidden="true" tabindex="-1"></a> <span class="at">data =</span> district,</span>
-<span id="cb9-10"><a href="#cb9-10" aria-hidden="true" tabindex="-1"></a> <span class="at">model =</span> <span class="st">'poisson'</span>,</span>
-<span id="cb9-11"><a href="#cb9-11" aria-hidden="true" tabindex="-1"></a> <span class="at">R =</span> <span class="dv">499</span>,</span>
-<span id="cb9-12"><a href="#cb9-12" aria-hidden="true" tabindex="-1"></a> <span class="at">listw =</span> q_listw,</span>
-<span id="cb9-13"><a href="#cb9-13" aria-hidden="true" tabindex="-1"></a> <span class="at">n =</span> <span class="fu">length</span>(district<span class="sc">$</span>cases), <span class="co"># number of regions</span></span>
-<span id="cb9-14"><a href="#cb9-14" aria-hidden="true" tabindex="-1"></a> <span class="at">S0 =</span> <span class="fu">Szero</span>(q_listw)) <span class="co"># Global sum of weights</span></span>
-<span id="cb9-15"><a href="#cb9-15" aria-hidden="true" tabindex="-1"></a><span class="fu">print</span>(m_test)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
+<div class="sourceCode cell-code" id="cb9"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a><span class="co">#install.packages("spdep")</span></span>
+<span id="cb9-2"><a href="#cb9-2" aria-hidden="true" tabindex="-1"></a><span class="co">#install.packages("DCluster")</span></span>
+<span id="cb9-3"><a href="#cb9-3" aria-hidden="true" tabindex="-1"></a><span class="fu">library</span>(spdep) <span class="co"># Functions for creating spatial weight, spatial analysis</span></span>
+<span id="cb9-4"><a href="#cb9-4" aria-hidden="true" tabindex="-1"></a><span class="fu">library</span>(DCluster) <span class="co"># Package with functions for spatial cluster analysis</span></span>
+<span id="cb9-5"><a href="#cb9-5" aria-hidden="true" tabindex="-1"></a></span>
+<span id="cb9-6"><a href="#cb9-6" aria-hidden="true" tabindex="-1"></a>queen_nb <span class="ot"><-</span> <span class="fu">poly2nb</span>(district) <span class="co"># Neighbors according to queen case</span></span>
+<span id="cb9-7"><a href="#cb9-7" aria-hidden="true" tabindex="-1"></a>q_listw <span class="ot"><-</span> <span class="fu">nb2listw</span>(queen_nb, <span class="at">style =</span> <span class="st">'W'</span>) <span class="co"># row-standardized weights</span></span>
+<span id="cb9-8"><a href="#cb9-8" aria-hidden="true" tabindex="-1"></a></span>
+<span id="cb9-9"><a href="#cb9-9" aria-hidden="true" tabindex="-1"></a><span class="co"># Moran's I test</span></span>
+<span id="cb9-10"><a href="#cb9-10" aria-hidden="true" tabindex="-1"></a>m_test <span class="ot"><-</span> <span class="fu">moranI.test</span>(cases <span class="sc">~</span> <span class="fu">offset</span>(<span class="fu">log</span>(expected)), </span>
+<span id="cb9-11"><a href="#cb9-11" aria-hidden="true" tabindex="-1"></a> <span class="at">data =</span> district,</span>
+<span id="cb9-12"><a href="#cb9-12" aria-hidden="true" tabindex="-1"></a> <span class="at">model =</span> <span class="st">'poisson'</span>,</span>
+<span id="cb9-13"><a href="#cb9-13" aria-hidden="true" tabindex="-1"></a> <span class="at">R =</span> <span class="dv">499</span>,</span>
+<span id="cb9-14"><a href="#cb9-14" aria-hidden="true" tabindex="-1"></a> <span class="at">listw =</span> q_listw,</span>
+<span id="cb9-15"><a href="#cb9-15" aria-hidden="true" tabindex="-1"></a> <span class="at">n =</span> <span class="fu">length</span>(district<span class="sc">$</span>cases), <span class="co"># number of regions</span></span>
+<span id="cb9-16"><a href="#cb9-16" aria-hidden="true" tabindex="-1"></a> <span class="at">S0 =</span> <span class="fu">Szero</span>(q_listw)) <span class="co"># Global sum of weights</span></span>
+<span id="cb9-17"><a href="#cb9-17" aria-hidden="true" tabindex="-1"></a><span class="fu">print</span>(m_test)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="cell-output cell-output-stdout">
<pre class="code-out"><code>Moran's I test of spatial autocorrelation
@@ -486,14 +488,14 @@ Moran’s I test
Model used when sampling: Poisson
Number of simulations: 499
Statistic: 0.1566449
- p-value : 0.014 </code></pre>
+ p-value : 0.008 </code></pre>
</div>
<div class="sourceCode cell-code" id="cb11"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a><span class="fu">plot</span>(m_test)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="cell-output-display">
<p><img src="07-basic_statistics_files/figure-html/MoransI-1.png" class="img-fluid" width="768"></p>
</div>
</div>
-<p>The Moran’s statistics is here <span class="math inline">\(I =\)</span> 0.16. When comparing its value to the H0 distribution (built under 499 simulations), the probability of observing such a I value under the null hypothesis, i.e. the distribution of cases is spatially independent, is <span class="math inline">\(p_{value} =\)</span> 0.014. We therefore reject H0 with error risk of <span class="math inline">\(\alpha = 5\%\)</span>. The distribution of cases is therefore autocorrelated across districts in Cambodia.</p>
+<p>The Moran’s statistics is here <span class="math inline">\(I =\)</span> 0.16. When comparing its value to the H0 distribution (built under 499 simulations), the probability of observing such a I value under the null hypothesis, i.e. the distribution of cases is spatially independent, is <span class="math inline">\(p_{value} =\)</span> 0.008. We therefore reject H0 with error risk of <span class="math inline">\(\alpha = 5\%\)</span>. The distribution of cases is therefore autocorrelated across districts in Cambodia.</p>
</section>
<section id="the-local-morans-i-lisa-test" class="level4" data-number="6.2.2.2">
<h4 data-number="6.2.2.2" class="anchored" data-anchor-id="the-local-morans-i-lisa-test"><span class="header-section-number">6.2.2.2</span> The Local Moran’s I LISA test</h4>
@@ -629,7 +631,8 @@ Kulldorf test
<p>While we test the significance of a large number of observation windows, one can raise concern about multiple testing and Type I error. This approach however suggest that we are not interest in a set of signifiant cluster but only in a most-likely cluster. This <strong>a priori</strong> restriction eliminate concern for multpile comparison since the test is simplified to a statistically significance of one single most-likely cluster.</p>
<p>Because we tested all-possible locations and window-radius, we can also choose to look at secondary clusters. In this case, you should keep in mind that increasing the number of secondary cluster you select, increases the risk for Type I error.</p>
<div class="cell" data-nm="true">
-<div class="sourceCode cell-code" id="cb14"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb14-1"><a href="#cb14-1" aria-hidden="true" tabindex="-1"></a><span class="fu">library</span>(<span class="st">"SpatialEpi"</span>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
+<div class="sourceCode cell-code" id="cb14"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb14-1"><a href="#cb14-1" aria-hidden="true" tabindex="-1"></a><span class="co">#install.packages("SpatialEpi")</span></span>
+<span id="cb14-2"><a href="#cb14-2" aria-hidden="true" tabindex="-1"></a><span class="fu">library</span>(<span class="st">"SpatialEpi"</span>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
</div>
<p>The use of R spatial object is not implements in <code>kulldorff()</code> function. It uses instead matrix of xy coordinates that represents the centroids of the districts. A given district is included into the observed circular window if its centroids fall into the circle.</p>
<div class="cell" data-nm="true">
@@ -707,7 +710,7 @@ Kulldorf test
<span id="cb30-7"><a href="#cb30-7" aria-hidden="true" tabindex="-1"></a><span class="fu">print</span>(df_secondary_clusters)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="cell-output cell-output-stdout">
<pre class="code-out"><code> SMR number.of.cases expected.cases p.value
-1 3.767698 16 4.246625 0.014</code></pre>
+1 3.767698 16 4.246625 0.008</code></pre>
</div>
</div>
<p>We only have one secondary cluster composed of one district.</p>
diff --git a/public/07-basic_statistics_files/figure-html/LocalMoransI_plt-1.png b/public/07-basic_statistics_files/figure-html/LocalMoransI_plt-1.png
index bcb22f6f4b9785c517f18d28dc13daaffd6be770..4654d2bf6ba6f175849759bf47f7103805d12c06 100644
GIT binary patch
literal 44265
zcmcHgcUV);6F3TQLIR<OUX`X0I?@r8fQob?y-86(rHO(H(hgN=g3^l+MJWPGktQt`
zigZ+ZRS=LSMd^@x@bmrM=Y9XW&vT#W9>O^%yEC)1Gqba^vzsV_;YE5HZW;gpy^i*I
zV*n`OR|*n_0)Bj&($s|?WP-lw1^6L%bXm#C`O8^Z$yveoEIEJpQwJBp->u~QkBW~j
z_$ABA-`~m~zSmi0!JlCdAnXrg{sZ#2%JTP@JF553^3Q@lb^cky{&jW!b?|*F%StW_
z#`l-Y`UiJZpH&C{W!1s=;jH1htYP@RRp&2P=WkW#?++u_Wm(l_`Tq~34!(wAz`9}h
zzE!tXSGP4>x3x8#Wi_1TKb!><@IRsf1HoiqkfUo0{u|!ff{AX`Wx=3F7)J$1cyKY9
zOm-AGjD|%)?WJwy4ZyJnNB@ZUpKHzl@B^Lm8m7KkOAT%nlC`%^-t9H`GX9+_;^i;B
zN5`&?xEuCPmkMedgtzd0lJVCcelm9E#n9Y7dGbfyA~lg2?;^NX&@lM?+t!j;y%=q!
zxxiz5Lm=D~kdCVEqd$6S0Z911`W!m?Qjg<5dPvh?;5#V>`TtQB5k*4axyJXz;N~F!
zbbWpo0l-Ot1h~G2H%Q<>hXKU%Ta7dTI>iql=Krq%r)WqV!3T_!O-)VYjY)Fm0qN!>
zdEHysZ2ymvVv^8n-&Tv?zbCCM_PVZl*OR-|*O&bR72Hk@8#nHM+{<Z{N_t$ieP7t)
z)0Z8V|5%)&CUG<cl#-*>U(bA;`z{^itPykVmK-&SaU!JWlUfvbA=#Lt!&T-paq++2
zSt3`{8PtlGKF&E(Y{qfR#e?k!jrVgdfG??a|9^nW$kJ7EPVk1;;eNmR(wl~lD14X}
za<wOfaelP*cP0i>n`8SQ4;cK^=wW8p20g+zm@GZ%f0igsUY#a;-hMfWhqXG7j0F2j
zdk|Zr)l)nc+S+Q`h!6aGnBR1&_RE{+wLdwrFpxV53*DTc_^{Kd9=y}6@i2$Hvo^RG
z)uUNvz_{O~9(-JQf00|}Prst~YTZVs`hko1%4iSQ;jaBzzU}5q_si7Y{Ak+WkFgKx
zKA2WNJ{-KeN;vo`Eyb#{HgY8KAe-a;;SmQjsqwDC`}vdG{gf^{<4PhH!Dq?iRtBcN
zUeC_gzP|%=*^m70Or^xn4!vy-Ywwn8?LTol+k(gD=l_hO9l4r@z)Wb9OVp|Lv70@Q
z9zOdmc=~kr{tkz#^jS6Ci`0)UWlIC{^p@Rs&&<?=SaPNSP(KJ9V&AG*CdFTJ%-nyB
zT&?&@^{3)Dd3_@{qsEJ@8Fl;P;`TbD-i1EroEjSZ$PY(ni%_GVYOfWmq93dTvH@^4
z2|Xouk<gg^0!Z{Fh*0_Y_>^~<|ET^=e#g5jfK#;3wpfc&=&+c0`~>*EeC?saSR4^J
z&9?be4+Gs>&Frx#v+x!jt0O!g+dln+QtSA$7}zk5o81-IcIrO(aHVK_e!>C)WVEb6
zWAyMJ2UlxwHnBU0d@z@zez2U<m*1>Tp8amR9W=e)y0o4X#Kd8P>*0X-$SaE%=Xb5&
z8z0VnIo`~Hp7QQjKRg&*L_TPcR6iWwT{-+{KE#(=lRqgktH!?7^POThVE5id1=H`N
z5yn5Y25+(QWvr1dfO2+5v6zo)i7Hz~amCC;e|t0_ac&}4OIo97qn2LWX8)6(mZCpt
z7bBHAM@?#Jn=D@P=`>T*;zlF3vz&@t_S|lMIc&9{n)VRlw+0M+y&oyw{@3AGWAmrY
zZ99FZ1c6G-gXCsZPN&}^EV2moUxz96fRuEwlsXbKxzMxqBkWw&x$~d=Yt@WEBQ0~#
zVh@@XMVcPDdGB`{V{0Ml=X%;7LmJYpKi?4}MYXJkk()V3ov@=gwXrbI1U%OCoIhcp
z?%JnDHq9CUOdXc;AtWBn@R4)Ke=03V_n2;aNOw=#S-rHid-db#wC$%CXWtxKb+sGS
z#Ewy@@9-VQ|DKdXmd@Fz;j>>yMA%Eq<F=SL-zuocPb0)yqTX&-%7#w;%F4&T5HhaY
zG6Nv?eFlo1n)GAU?x8cSgBqIDy4xMJCZ5#Mfgw$FU*RlB_3H@qXs<dLKbx3|h2p+*
zF!q{vRjv~MxYfP@M`GbdCzZlv+gpuh)ojlX)=pCVH>Pp4q~?RY3n>r6o=)F2`ZLQl
zNR2nOdvzuLtP0YrbSGo`3u{v?7eYC0ooWuQsj)wONlFrwaUJk5Q{zAU7;Kk9CbUx{
z0AvAOIYzqnWc%t}fi&;ngX-f&Sx9XB311{OdGH#l^pVBk-<-o9@^Vv87s8z!u=C)g
z(p3~bS&tU)uGMs7P=L#M&CU6n%S|Ukc!H=$&WMLpb9EJifEJ^}kCTmj9r21Bg!*hq
z5(N1Tcmw+e6K^od8&gxBNXlfKFXDf=Zd`NknQ5&K0@LZsU&;2V8Rm_-F3`^V3IhOk
zi1h}jk>;`>qWbCnQdzgfTy*|Jg947uD}JJg)%G{hNNdh<@q?mb+WE(1X?3!q7m;uY
zZ*#+`HVN2v`NCczt2wt_Fw|(aUJG%sw5KOw`Jb0nt_um8Ruwp$-Scu@s#d+Y|A(>e
zYOoRS<Mz2LhwCx9pCFDitQ{RKIYA9CS>8}6GfMn@`cggPF&ot5OrBHc)q8BR2&9lx
z$$F1~l#3d^go0l(s3g{s3O`x29N?6U{ZC5{!Hf>W((dEa5<@(5&lrOD?_Zwq5BE6U
zVN_k%cd|7sl~2!Z+$iYp@wF>$MLX-|{sc9(<H3@bsq0(0&kWR`9hmm~wvBUpdLqy?
zi4|(@Yc#)Df|{V29Mw(Eq@$qk^o8gsPD%z9E-6n(ggjBnO(xZuA^)2?!~R|<mELOT
zV&C_33|;g*Mc!d_9GX4uS04G?*QSg2Yeo`a?a(ti>$HWuo#3C+3`Q^hY1CgzxB
z{SkhRX?nFgGQC@A{$Cx5WzEvb$FBX5=~bTlyJBZie&^rh9JXUHW&iX%Q!vYRkz~SZ
z-P5#2<tnAUjW^?Kd#cjEx`mw*bPDS`m%Kfd2#k7l(sd!nsRpW+E%v7;$pi6QYPT>X
zi&n!M>Erl#+LP@D;UQYEAj3oQB#v9$%~JV*XHkiVz0Hr@|GbSF!goo$sTj9Ach&5R
z&O`N`{;l|u-)FAu7BvOFICJ+S`3KY0_12$n>?v)=zFvxqSik-K;^-ETl1m8(yf<
z^1DMjhx>}}Z|~ZZ4;MXz<t4#2)iAVW(D%a#yAFd3hzl-b8W#{RC19_crSc}S+b=u5
z_{X#V8n;E%usuv*Bk}&HR$oDc^2gtLIc8K6z5k+<*g0m1kMC@14naU1QhE1p=&zke
zPTdR$L*V`YQ^<4_%He>ObGwP-fA#vvsi5_a;I*(=DDOXS*J45qdjA(V{67RiID{`1
zJS%6@9?i`V2O0%GrY;~5@Wb~K9=-!R*8g`^f00wRO#??^o=Yhf;NT~T{*okMI4bed
zOTz$G_+=`nfDKR`wN@kIjsi#FyqBse4HO2yfH77rBtRe4@hqT7u%iWFg+?3!9Z?vg
zGeb`C9~D(cqyzjBg_Q_ygx5b__dXZ`;=el02%!8g;EZ#`0RmhEscJ|Ajzc&OhG|fz
z{Et4<Hza;l{a@d;WZ3>O%YqzBg3N*T;e*ErkcM5M_;db)IdHfFKwj9MzN3Q(jV?la
zRb8JBklcM&$y9Aop=E&727rE~8NeH&e2@t<!9$z?8ioY$G_V|mD1ZzHiL6tA;{Ql4
zl?A{XflNTbZ9@4iAIMn1e3}Dho^)nPks%C;g<$?x>%COT=Ku~5zx&$NBz__Si+}IN
z$~OT(5CPolDXucDm%jUqh?=8_nk4rDO-H#aa!ClND{nR5<O|;a<+k;cZ?}pEj`zb7
z$luA&E{TGbqE@~mLCqp50q-{&Xm9h_5wZ_vj64<0-4DktzXjlG{RbuhPu4M|h4v_q
z>iJ)^jWYryS>%#5$SLH5i+bXRtM*1dsQqH!F##L$ey2@oNWoi+t=OZ;--uCZFwb@3
zIHPGIKahZHEOKheF_W4w)5E7mvrk8;Oz>l_KM^*1XpQhCkH12~Ecv>0RuyK@avgrb
z#RAI$cBp6O_u(~IgK7$6?xz%9-m&9EloBMt@5yD8_i(AmShx?j<R9j2O7m~DB>Nx+
zcqY0QnVDd4K?qXGP!n1}JH$~&z`T14J}2wU%X=nuWM@BXHu-=DLHi1@@Uz)04*87o
zyDy`rymPav#po!&aXOSkI#nc$Wb&h({Bh?`*>24B{^{U46dw4NVxfyT+z}$PHw~WK
zZQ3QPDw|2+NIV-?8&x29#`<KB<e8j^IvPX#F~h=$hgMN)@LKCgbJ`Bxg+_V67`DF`
z|LPId%N5>Hwe^Xhv+TB?@evHM%!B;vN=}D^ZZZw2x5bd*$bRLG^%tuewng07S@HiB
zJVBN|qK;%L(npZUTt|4?MC?sDS1drtP#hT*tFGy2qAf>$BFn{)P_j1!GZ+)PhEz`m
z1h;yAaOYpa2e!{;ZI#D{&3F$O5e`R6lcur2%sbV)odcdn?bPE^3>4&o_$hyC{;xnd
z+TL%D;HiE;0=xxqLv}1D5e{8do3yC^yC5H+cy)q}Pj<bDL}g0SAbXKX&HS)A?|Wq<
zYo>>o+&`ZjldLp|`O*xLGa*^7BP!r{zEr0;Lz#WWNT2(CUF$*lx|B1LxI_&AQ7n3r
zhAwS!efNG95MuVJg8lM|!jrpI&!%TJhUt*2<(BNDs^qPFqm{KBqn=zukN3&s0>OZ2
zxABGMH(r>RtoTXOu|+jprUvZF#J}#@l9fxRv*V5RwTcq`30Lzp$9dGds`gs1zNMLZ
z{i2~Uu`RQl?B@7Ngwk0RLlVCd*f&`dl;1TqHAT6A{n8-BIcw(FLyoy|xho5t1bpw(
zqa;~8?;KBB<_FSLb8><Bt!h>v{i<0c<ZD-8uG@`BMPw<D<yFUfeRIw(nj#?c_LSkA
z#xMhN)w-sfY`Arp&g0kJGk4ba7=5VSBc^Vm;-iZuh+gb(vuyEtQ@!~Qd$f~w!njeG
zeOil6dp7>0h#>K^MYP+>6ZoF^kroZ9OcMQG-e+HVc2H(@;7r{w8PK08M(M1EA-RRH
zLq^m&gpC2yv-P8TO&er8aunAOFP0`@o40IW?Xu5%u)KDhAg6bbQEiH7_r*eknwkKY
zL)fdTfg}B9<Xr|oUm#1z#N53|kK#rlA}HAvMH<hX9&9S#97G?ru_1)n%!wiwW!rP~
z-A@txF)7``6}MyWOB?rdzTY$Nc+y{f(8Dly){n)b7(r4LAA9F0wyAu3u&k|Wc1NKm
zSAk3`Vxgb8Az0muN;=>NMvpuKe;<mNfRqIOy^9_Bwyyr=_ZpYJmAB2x-={vjeir!p
zCl2+Mt4QA>7vs6G$kj2g#r>6ct~nj#d^*$Cfe3#CiTI_%j^Jn`t0{U6T|LI1cbAr+
zO@;8Cw&}YIdVAfCsM;YcbggCmc(7M(0y)*aPX@1_+MVh1*Zj~HsW9&6YruDt3A=oi
zk)ifc^1i}&(5nFxQV}Tax;~V_fxW3m5Do`wa^~xkZMNOyJ!Mk9*R#4o`;}kKON5^C
zduoFFi;pN)1+l7p9*Q(qio__~(<lR`84#{p0&#ANioX)WWCc)DQP`Vv^~KBm%uVWa
zL-`RejV>$qhzI<Uu0XF?mr{G`3P~ls_CA(%&!FpbaN*)|Tp{Ri3UTf-pxb*()W;xv
zkxTa5+i#cW>Koo#H11v$^Qy1!yV>b^^P-oRzhU9e+Spcq9?W);*4F$n$4fi+GbV|`
z50#pRSGpZvZ83D`R7yPtX#-;|<vld*;aF&IdZ<<X{Wv?aG#c!TdCaW5W~q7q@pZuZ
zWkPI>`JMEsZ22Gimu0eICO6HV6uq6~u|UM4r?Qnid2u|jbBH(giBAUsG`4^2rn>L;
z-aAvbsP<-E;HG-S=ki;b{uW_@i!sUb6?)Hjc)U2E=a1dd4j=qwm2g5Z?t_HDj`vSD
zk=TU5?TldNFF$YfpPg+FSX%FRTS@1Ymz%jA*HrfO%McpkMV{}oiA2gw9!o!##3lgl
zeQckrQ_VUzdiN}Go7z@{FRiC&b?0?g>l>nzmtFY^XX`f8Z8#IMb}-1+!9?;=a=K42
z4~J)dJ7x0e;f`q!adgeudZhYBF_IAQI2^EgVZbxs7CwK<Ivm8JoG)NVaiqEZhc%hh
z(2U=5R=mB&=Y@Kw@l(&h_E)gTUh*DQR>21ST4`G%7Gn@KjQh@)Wq&9dL)TbAH!CTl
zv{f?2fjFoh`UX504365lm+yalFGOeR7BLovX8NY)`2q-`96~vXI%rbBnQ-oGP%n9N
z=Xd7s2S72DldAn878)f+HVc4XG|^H-EA-R>^S(Of)Nq|VdGs|!+G=(uU3(-J+F>eQ
zG&8R5x;^zOztT-nhu)7{M<O^Al}LksLM~(W<;SsTg1iUx_cB?H|CZkEJXtK%Ds(72
z<5`^*uYxB1WdgCliRg|&r14a~IaIxK4%|+j@dVPtIDJu|;((H2NTL=1CLcZNQ*{K)
z7~NWJzv#*;k=DM%ePI0xXLS4_*q|L;(=M!zCjHky$q8?Yv|o|mJ(1A(R7=QFwM!a$
z^&D8QJaH4B<axHG_9?IoVsdC4J*9BOolA@DKDi#5;Piyn%HX1k+&1I{qw;J0L%pKk
zLdHVWqSugze8S-Cc<g75VJuQcx|ml<_@%^2voj^fyhRtDse5a=2M^Y9%$%Yi#{}YC
z`X5ytcHxg8#od3XtB+p>Eewd){YF}+u>hh!5}cLeT+J!5IV%g~oBkIoNTu)moib>d
zPAd8DS^MAq34|RBMZ5oSW!?fv5<(c^i8vZ8qTiR)@K|83@W&%9h+!%@5YGuP8eB|B
znUe!O3nQgfI-+iu%;Mb!-PiB8)sgg+SWR#wP6VgNZe571Aj&~=i3dCS{{s~rEft_?
zafWBPlOLpaP4a^c=pPs_4>q&3>VkmH5wTX=)8XMy_D9;C>g7I4Kwb<aZG><g(?-}4
zsGvI!nJ@mq=d(mgXI^KsGp}KRe3#RG_v4rEwR+@3G>P~*4S@}lg{|#4R_+nyi_7+(
ztjzl<Mo*<O7wxAsRw^G`i{+!xMw2uVd3E$K(IZGa_#q)wP-6Kjpwn9&Awz(6JU{*+
zcEWaodK9o+oOe_2@V7dWv)lbdUT*gHbD$QO8d2!qWAOy-k6_{A&oWi+dsL-4NR})_
z_0Mp1xy$DX<p`WRImti16Wr+UMM1b54)jAm3E5tp%KM{!g~7oPo%T3jeS^J({)-J5
zOWJa{OMq8iogv>SF6n%)qFG%^B)RLipZ$2ASPg2nbbgb@i2BrsG>XquSMmf%Ceo-x
z2fJOaD^ks@_+%tM(v+$_1q-Pp1l%clzB*b53Ex3>>EW-Uh#MzX>t}O#k|YH{BJneJ
ziW=q6eG1C#r@qC(fz<*O_+e(G47IGd(cf?n?L$33Hb##K&{+WZJPp~oor+%5$mIXL
z3EI=JkWNBgtpx{Ub#m&X&Us>@!<aVsCO0q@d=ht669{WiAWAQzR&Qj1&Q|~_wHtuG
z%sO=zNj#2n07SI20fxleM#!=Q5}$)b34ci|sh1M{+M^&wMJ^BNy7->`w<{NO>H3V|
zD9@TO0Lv<5N&Iz6TaFIMGXn$TXP8wVk7|E?tTp`X0gy1FrUb48#RwQ`Kau$7ry3kq
zE}*XTK5lB-(slfJ+F0-0^=<nBYmuzG2tk4t+#l#m7||GPh=LjjH$xqjvEa_7d0HQr
zDNQ-5o;cX))5F+ajD@%e0nGAFo(&k{?VONzol0!r{)46zmJj#1k=i5o-}M1WTpc`s
z@jE&M9Ms&38rvBRt3oj@{iUJ&>G6mzJ^!rO@RL?g1(bu*=AY*Za$~S{NuV4ky|Tlo
z(81Q^9(}}E$fxhqv9#Hm^selO`jUVejf+i7VT_Mk&@<@$LhY0Hkrz6i<SYQ>pab5T
zK;b3LkqZdSM$ONM7UelIzOyO%o(15W=_m!O2;2FNhBwToUQ4H+W5axM^Kj6|LF`F4
z1ON-ZmOoMid{3Q?w_n@)CCu>W+Gjz7UImM%x9%g9^X|kun)q3hZw-#`Xu9R2w=yVQ
zXz<&#%tQ)jYYfS&z~U1J8anKBdPh3HU~<NXzNAA<{or@`yRpxA3O=QM8&<gDDA!>X
z!V|~}se*^WHt7q^_+fUW3~$MHlay;>@ooI9>MEJi(R5Em*e3a-7q)lmhqA?qpTT*}
z4^N4nz*Dr>V4*%j6IT;FK|Lh@y_&TZ7}>0gS~-PE=m9woh=7LB5&>dc0x1C&l7D5c
zgQ{3lq>=gHsfj3<3RVs=B^J{nIO#EX#jpD&q%Vc+T7HxbuJ_SXRWhd}&2By)q;S^3
zkgm(KKq6j<lE$?R>(Shh?vX!5E0KwDz;&BX2(aPwG=Wy`Q;;T(zX{7k$8EtQ&I(N=
z<M4{WAw`D~#7g|Rax~&N>7TlVSMqMpBB!K+^&e7tROD8_j$Bj$5b`=c9UeQJaqJ;y
zl0l~&<->*p18z!$a)&;zVnwfBi0ny}gS?LrNCBq5r?>`3lq!VCybnB-f_q+gWH?u!
zDAw69E@59&Hv+~D97rP4Sr_Jd9+y!B@^YbDNSQDsuTuLSNiQ?H^PPF(>`CtN$K+?=
z^pT&4)zafjDB~-51Kw^`ws~<SymkG0j006iSwxbB7s<WQQQOMak3hB-5avXNQ!ros
zZB6KFd-T$yB7w|BlVAuA2Sdt_fcQZcoi{apbe40?=3=74Hm~KZ+P3{ace2)dzAYaG
zDLb}O2%#P7k$>Eve%(^fWsg-+$ae7*viBbadDGY7C&(<<JjFojpIkQL9Brje)52HN
z{9JM<Nc#4kgnvWTo`i)S6I7qD93?ZldX@6ZBfl3biL_Uld_F{@al8MN4)=>L>`>G|
z6*`OY@MTAE@=|}RUKc~m54O|n8r`I7^^@X<Z6gxye(2#T14E#?`<6@K4=WJcp{2$*
zt9o8MeDjR_4G!^=waG$RE{s3D!^&ar&bovq_09cOL24&_l%b!uFMBd@KkYjC&n*@N
zR^iG)$L8k%>!WCiPP6%HMh4J+8l1sIx;=Ks=hoie&DVrxHxJ1Qo5qxFZyKVjIxB>3
z_rXE}1^4xDAiO@Kj81~qoklSkrx}6S%9aw_sHVd%^*7*EN~N!M4nxvUen3I!uuVeN
z%<N>ogOq>m-GJxEgmvptb8cpJlmiOap#k3IX1~=%Vufq+f!Ei2Umy){M$Id35};+q
zh~Y<)zfhH*X%5{HsRqZj1op6lr<Q;>Y-X1Ra!@2#zU7^s&A#w$V1PL?JIDKQ^Zxr+
zd6oCAb7=k{^qmJkG@H*s-md*@OfI4?PP--8UCn(f7E6myMypIXVz7VcbAJY50mBg^
zr$5e~lw4nDiOD;^-_&w<DOko>&c)%Xz?9-uBhhk`MGevr8)m<|zeYe2iA~`eeCB5L
zQDoB(Z^eM@<<jaYQ2m@Ud*v|_CD)cG@k2Hwv-?<?nrABiO@6`Rf|9LV=kGs9hWGD=
zG2xerV;6fXi+>93$rB~05c9Nk37tkf7Q(tooyhZ(wF<7^&Ht#N9@-s@1<kV}q!a7e
zRygr#G4Cx8IdJf5Tsy3s7ey435HfF<N-K`R;R;vv2lz!LUn+`zL{uZr&tEe%()Wrn
zo0X%mK&*&>$LB89sazSQ+S2*QD&x|Qv0aX|@Dx3wSa6N`rAScnLcyT!m--U-@#AXI
znZW{hM=Ir?XgpB<$K3;`Idf_aE5LgVg@5WM1%B#P@=82L$G9^fhHM<qf>qkV+QWhR
zxSa8P;H-?ns}Hpd%YD4FE(D<{xbMGuE&1aDx59~W8Re)j7X!zS4Li$L_Y;NoHNVnA
z?x%HoW!$N)DUa}!O0OIB{!IUDs|9SIr@0_YtF$w}Vy-;lk^*JFe`?(GJQ&p3v%Gwc
zdm}i4%!DMMoE2fDSSBdGlgE8FmME|@LNy~VUHbqyUF2s?UJ1Ydh<c*>MqWyZyA*J}
zdJaSrJ1G(Kks8PZzryyNt7ys310`Ks#uOl9`ef|8D9CMei~)8Z*Qwgwv5;r!$^jD1
z&*(DAR)X>UK<e0`1X_Fp<G-)#t!1T<1%NZbB@Fm{ltBzpJ6ieGt!xt>m7f|Zf9~BV
ze4&~HxS4whBW5KVOo=x~=I8u_j-dY_i`<lW5jB5<LzT1c>({L7PP||^PUN<J$woBO
z^Jo$KwJx5axQoc)OB#mF-QoD2AUC2-gvtPg&#B;wh?2vD`rFAgt8Lc~DOnZW|7-+O
z5F(Grs=6n3^amgal11D&K!4Rr*a!4CSZw`4oEa%}edTt1zw7{+3MZ=moduvhyavg_
z2E>61=sGwqSfO?-^ARfudsPyx=%Ke@C{F+#F8*=f)`FeukwT;M`FrgD%m969a8?vl
z<N8e$3_%8ExsqXyWXH+RrCMNqoJ2p`K>vB5-BdEk^rhR>|BVw)+4=Dp0^|0&!-ddt
zo&tCuGS08^Voj;>dnx|Oq9Y>ye=jLAF%$mjUj!%2_tolr=C@c@$nh#;=JU+b`5mV4
z<c7;HX>mWz%3bgED*K)DPv@oiPxX95wci6Ls<Xr#%qCtYq|l~w=XG<i!8qR3@r9WP
z?YCd@)>2FPZe6XrA;n5WwYy>=snXSm;-*jbzqxs$)m?XopF7pvHLWNJb)dI4q6}0t
zhu23VA+kIuBGoJV598dOetqF7NSW)H-rGuYdr`hFI>M&Xxq~CKBIge0c##gwM7nlQ
zEOfkp1ynx{{uI}#0s<u(rLKA`S9}6iyTPw+*fqaaut?}DW)Omm=jyW2WVxeRvlo$U
z$GLvXVWWb$zRH7Rry=5=XM%@1T9@5a5p8zo?mqa1Tu*Kv!O0D5<^*6q{T-UWw$DI}
zrAAz5Y_O5r773ty5h26eB6<;nAC2Rv|8UvS{>2_67@;=OLo5H*cDX%$n`XZU0mq$q
z|AGe5N9%hxdq5M`#q3U{shOvZ$Q!SV$34D0xr8LFrH5#ie3)s!;h_6;P>2&zfVbj7
z_VS*Hkjp{u(pPHOqLZ&fz=sE*!a`a+xP(F60{fzF*O=-~3^z-BP$wK5k71$N0xX0I
zN0J<5Wg{*W?S(^av_=QVTx4h>Jn$dfT|Ef3d)qhz7HykfO3H_UjTC<3`O_RitLhX<
zwF00-gVCzUq32a125)j`lo(`<a9^mLb3jy%3<iBU^~U}2BfIfEzE+>RKD1$cxqKU{
zP`MI{s1QjDMD7SNCOj`(&cYi?0WC$D{He5<J@XZV99E&oi5DymPkJS8>g><w@52^M
zgSZ}hrnrk);jCFuh!6_O-l)>Z*x;eXtD-q&!9zC#G19xPh+fl(uWykKY=AcDf&hrS
zWq}}g3U=XYHP~;}axip!;vKMX?|CoJw)JYy^^9QG=CRYy3SUr<RUK!hA;BI2gLjL)
zi};lX#ckXF?7NW*cJ3X13x#R}@!lBH_uMVx;%xuIr|l}QO+K~4UHP-4!Nm$oJ1iHH
zrvM!uUaWq_Vn!Ug78=;zF(knh(pm(@v$V`@sZB}0)DwR^!{_q-{j8BrkL+bVzZ!}_
zHJ4>;!h75_mvtmjP<7P8-<l}e>uEsVh9;}2*V}VI-ZtdaSqL?QPp5-XT2GjjZP4=n
z2)^Z)`-_|#h4i8XNtLtN6u8%@mCMV{Kc7pWOqr4AMYs+~M)d7huu$ac>9;RYoM%l2
zk@OAfyIHd|*!Ww@QL>^D$x6?^^Rg(FSSnLp7)j1>5{m<tm2pAya6VNAgIBekC{9$3
zk8(I>$pmLF=@1LLr&0hrxtbfqj?#H*mkhetG9)l*z6@$zW=?=k;Jh_}wSbC=c_GED
zoSW(Ohz)Yl4bSCI7r$C1#!}^8kc$jpbUA)#)M!_l=UgAV#C=hHw<1)Q9Yv&$9ygOj
zL=U~Y%`BF>kz-y+XWA1`%?jBL>%YVfoIQ`o^8%M6Q={aXyHBqBKML9xK|70M@D!`O
zzs2qfv~<68ZD%;{>9?P;RYoDFl0L~Q`r~t6-QOVM@>xp$I-lP?%8P5(m#W^De(9r=
z5y6>gfX?^VJ}+_Tgvz(}OyepZ%|BiAS_r>U6H#V0eu1p(C=Dh8)GDqtYf=6h;VE0Q
zIenty=AB=9>K+N@arf&+d=~n6|DzMD5vf$5x~1`gqj0LxUNvL)i)!wvXo;LRY*6Gm
z=!5=r?OW64d%r??m~Ov{id#t3`|<sPzOX~GV6l%py*_ojF&3B<H;hLUI#{U7xh96V
z-X2`7zw))rzS=wDtua{Xm^CCS{Y4tcQq;>g@7>&0E5Zgn&5d%<C8UG`PQff$XA?KE
z$Dj2zT_ivB;s4qk3zW6UCo*`n*!sCQ|EX4xx$!v>1GTX0DHU-8T+6ZCS_dscWGF}r
zS$L^0ZSX@`#QO$k!oznark^^eKG0@pfo;Qz_@S#;%b?shKwlmVahnQ(=%;W-357eS
z0b=o6Uc%G8D513Zu7tW6azVlMHzf}Mr#J}qY<YZqPtM`ODCCkgb_t02$fTRiwm4ZR
zWnhI-&LS9y`jy=eC6QkWx{nfTlal0iMDI7QtIEo9jF*djxz@mX((8_gTF#WZ#TxTi
zbkbgbztbJqLvY|!HGl+R<L<P`>5tEy?H(-g)NYWOu5p-~30!%SaiV$q@tIiBsd3v)
z`JwAKXFV1AwVGY|n4gmAX}l25!Wal4%V6l3)t!D})7(zYYJH^CjoEI&5<w{TRjw4w
zMYLZCFwm~GG3YN6{34Ff<ua>`bBWl&r&A*d@v;Z(@5FAkJtA|3WIun_QhVmS<wfVu
zReq`a(SFJNkix~>R|k(XN^if3UZA*QQX$hEDgYFTohU?smO`yv;(Y38M^B8cLB8r|
z?=2^k3lY3H5%pb4HE5RlQe)6_wB)@Eq#`ErPL%(M7Q6?_K`;*m-<}^XXJDb*!a=T{
zX$8c*r+db0dYvh2VqcvN>t4zFJ=ZXJtR2DKdjsQO7pK9x)avSM4KivV{{Mp>9W;0%
zRwzr%^!*FL!+4g|0e(z-RhOi1mpY@a9hu_xjMQFTKH!vWzuSNYG;orM8AW9NxmFYq
zF1{tnyU}urp6y;D_L~6PD{DT(3+6XI7HprSH)CI@be=8wl2<uYzaPAP`Op1VX%<#L
zuY0D^Cq3o|$qw~XG4@`p$0p)QB2|n(2PJit3&BIKh^)v#W|ur`J=e3Y?v^p%AL*Z;
zAzfTgP`)5IdHr@8T~Zv3JLO~e%ce$ZLE!$zq>HOwt?A1R*}?J$rzh-mMX6iE*@XO^
zM{aq)JFADlaX+CKyDOoV__?UWTV6$<vRw-cSr=sO?|jwjc{+Y}@g0B9LI=U{u#DNS
zkeTb-`3&ml@@bbh@9y47V+1}dc38kePz(j8xx4gv$-wfA%?N)_I+|qUA_;ElOzKb}
zN}nR8(#yL?u#g?m9fgp=m*hVE{=_`wlA+m?C*YeQa^BYQOZaCxa~iMi&+=HnQU2D0
z9**lKMvoMKgaK88p&h7Q$JIEkOXh1WeV*pJPFV26Ckt(ILuR;&k6AjB{+V%lgJU6C
zC8A;H!2OVNCKL(0v{dmkjb4h8Wn)ufgP3UFpQyD}Ja;DjbnvP4j^C2cN`6eUBeAuG
zoZhR?QH24yQI4`lJsBd_TnuW=f936>=PY3m&F_vkL0?}4)+Y8J$%7xWoJqG&dk~zy
ztyVQJJ`Vi#0QexCg)k7CD=RS>u$r_nV!CfLTV6|^eQsXuahAuwF%l#+#rziipPYsS
zzCQkY#+FPcyWPZ_cux+Ia7|F(t2aj_hdXsJgS<zLcUIDK;WeQ44v~dBfh_9b&jp3r
zA3bW#Sr7!j?DNQ8PwIHZB26HbL#I5{lVNB>i{C&*a(|N%uR;4UL0||vLIkZ7X*$n)
zcn!Co(ORkCx(yOnZfY#w%RyS+4u32^VMC_3#rMVdeElJgPnizPU%nllm}9YV-8GFS
zDS{tRu5O3Thv^)5iP)B>qj;X9r}1+ch37ha(e}1{HYgs+{L~Jtr~(Yb-n!m{6Al;$
zWV<F7y1$(}?8edg$Iy<?<)!hA;k%dhtO`O<G{5;@DSWg9)#&BCz-t^(o*qb{@wpe`
zXCoy91TXZdnik)eSLsBx<6#mrDB?t&2?tfut>W37y5+J&MgdC^S_16@LA^+FCv|^K
z&D3dl;Wk|1D0@)!2{w4@q!4~!RkL-Qv3*56WwGWAiulg<FPaVuY!0jAQ69BM=MsDH
z3!W5PRqk&%i>;?nvcQ=SgJ<a2mdg=Tb<5a##!2wHVCPvjB0et-GHlTkK6x+3y>hE`
znkekwNeBF9hy;I$OB9GU((==9JEy1WDR!Qg&`fk3M-upK@*BeEus{zjbIj=czIqwp
zEX1%9xs&|h1;wX5_cLP}?^~;I_qTp<*b0ymPd%`%6G<@es(kdbDBGVcwMa}Xk$CKB
zM)J2P@-=1v2Ll@ERLF!6$*L;yOZ41TO}aPUBokuaB!Wyd<!LrQZiLL$L9Ltx*D?>L
zOL6QT*Pf_wU6lJQPb{WDXun@*zR;HaYlzqEr*Lzjx$GZy@RKv@T$s6!t7&&;Pv`Ow
zx$8VA-sVFR7~4<7QR!))v5tL>iNUL?sDxy#RFhijsvCA5Q`iTTCG+13vXtV}ZN89A
zk%UVOtPT{;ycj%k8O_h#B3I>UZt5uyM@^TM{3M_5UKa&ghU_1QQJ40_7YYvwcu)?)
zgq9F+!a*4PlDYod$WG$z)2ruAlkUc_voOk<P=9lKkgk7SGoL-8{(?9(Du!QQYzd$k
zbXF3jqA3k6r$_Ne?55^4?42~KS)Gi29<Jgv*g}vAv!*eVUyTo3-I)w^=_(izeN8G*
zjEGn)&tyd*EH#EHkqPHTL=wV<<4~kXGFR8Bd;DpJkAuB_eT{UAE7&dn)0;P>A9(t!
z`2^S3P$HnkeZ_+>rHk*l_@ZNDnhl;@;9iTFN;c^@cK*a`4}&chXz|O(SsJ38$8t%T
z1QoIrl7$%y8=#q5s<C`>4kx65Bq%X;O7$ggN*b8Ge}T`ma!8bYJ*cvS1wzmbd^*gJ
z#$uF%UqXaTV5n5w!c0=GlH=`P@9$V!sy#U^ds}5yW1|1On;_VY=^eZwHQk^W3M7d7
zu$et8L$<-4tm`t(F9S*4s^s{*eR7p9Uh*rqjDia8m2{P<TUST**`mvrun;=w20u{Y
zq=B02YPv8a_iyi+K?!yGsOQq2)USmpld0+8X*ATB)8Cx3XqLWJucrtB)6k>7P<0HZ
z25xP%-4T$?J6&t*^MpqOXG1DVN^V<YG1)=0Ue}oHN1w10dSzo?Y<X?+<*nV~yB9P7
zJ0ULw%s34BtDk)&J@zzMJi0*d^@Mp+AR%3>`-GUynRe`Bhcj^lvqlCd&8mxSE7((v
z5++#C2+Mym36UgApIOUoxpp8Bh<WXp)b9G1hB(^%*3nfuCl>X<2svp8=<YtjM1P&*
z?K<K8pVV_(brkbNZ`u(jCW1R8wEb0{zezb=&}G^AO`<HvNIxu@o1pXU7xBB*G!$!3
zltC7toVhW0%2TPT4Baak(O19Sy?P6AzO$y$e6~zk=X<X5e!Dxf|L`XXOI1uk=hMb-
zRu8~{&~*&{*pXGp7A%lXYz)&y*G9~!TZ)z>-Fc|n=j+xpz=T(&8rX%zcGTyTvwRd*
z<>Av)mUycpO}_-T_lti%=d?@H-^?%7oh{Q3jeZtm_CqK!OO}3;%%&G}dDD7sjJewu
z$e;?4&Rj5if5U&JI-C}o^$m;Ku4via)xMTW0Ndv{E`2=_wWW5Nz1WdJzA183!w!u6
zrg|n*UJFQ6@OF+Y7NVn@NNW%c$@oyPZiU1;_YH3TLUtdjwhOcsBk>K@(l<9A;PR^C
zFx$Z%p%R!%pOTUW{Bl^WH9Nd%4wciP_$g=+Lk&|37fu%8{02%U$fAejPqnz0K*#!(
zwc)co4Gyv$gR**v2vws*mZ*G*@(=8j!EsB(>F^ENb`$9x9)pu6|IuN|@iXfc$<NmR
zCAD+TlIlG<XZ_1L_dZe%5-VRRPHsXolkU#&Q476KDLy|@;hTCanA4&0H^u0^fS+Gc
z|A(0kov*@eHR!Zo(F%n9CgqTVPZOvRV+hWQXpEfEJIV4+X-!`080AZ5@a9GI`b;mU
zb@t0kg;<CR*8yt_0`5LS%usILHhr>v;fuiixBib@#aLTcPYi~y>hXJg?Wa9a;Zqm#
z#<f!^LpKi`$(}*sa!o2&RZl9qe@E9V*ciOd=DYRR%<XB1iuDtxC#PAKoq`jfdpjG-
zA>5lm+t6Dg9Xv3n;9SL03QONjgEel1b{|`v@kaQyM~C}4v}q-*&?2sAIT^FmX(UF8
z!c(`)z-n5N0uht-KN1;;CmC=fNTE8_;8%y&U&^SPYriKOP(LAdCe<C*``D@I3+W31
z4>tU*>5JZ<>HalF(gxgG2<I~;?8wzioT&4z2_-lNYihsX3YN(`W5K%??Wq$lXE^I6
zki)*>iQ-PwkF~)pHw7L{2!Jb=7}5u?zeZuzyNWo8YVM*6z$z}*iqdgmP}rX|10OoN
zZu>rS+|g6uM6RX?(^d=SxN-yJDgy%h5f2`%HoS7S%=SOAUUscRA&p{Nh7|-`EZ@KI
zZRS#0E<-tWS}`g~%Iw|Ssds4};#zkvID0Gc?nKMQ&OYJ7a~+L_N)Rr+D9g0^V){dg
zW7#i3!7Ye*S3NTw_?Eo-8Tol9=i!cp((8aiTePLc$-T4r1>XI1=V_K@(ON)^5E%kq
zT;gcY8ezyVc`>*0ERibb*wpEe3>y4=YJ18xCc(4SfN-zzTaD8qXX?AA4|8)Y^?tj)
zyD!IRc3(wvob2j*<$xinqVUlNI!yO^Dm$Ec1x}_W?BUG2Xfs6X#ye#A5c3&KDW}%x
zK;URJz2R-8#klYDjD=SU%$ZJF>^v{Hs}TC9EF9FI122ZS%tur@hH2KM>bsXUWw@W!
zF>f$Iuh<-!HW+~Ky--7q3Tf}Qv?jw4oU=Us@Q0-SWU<qKt=~m@y?93Xy*)pcHS9jZ
zEBy}0Md6iji>u#1bv8^Zqzl^Ly2qa>e8a>TF>{vxMx^=4b4R+<<gF+ze|^_S#ihJM
zzvR8i?0l8mzOWkEs7B$sH-g~odf3I<C};Ss`w*ER&#nckDV;enc;-OV)U@wC3k$7B
zD`78dTQzR=*&a$)KfL#3P|Qr&?L+G0Tw+%;je`Ktanl(U)H6&@=rzMYyh)C5pu&qJ
zh8I6r`K#?6_`)|ZUhqd#28*-P<W5i1>qo`5&pv2zvdL`vi#}0gv46@m6PMs-w<%?{
z8hp@YXqZB6@iI<0QS_e|JnuZ&&%kTi{#Yd|o7o{LEAUqjZae_ldpF-nD>lBM%$m9;
zJkr@-Yn?)Y1r48x{c^t?UfeMfQece}cs>?)zs5pWkku(};-EJ6mC2a-jdl5-!5dok
z^>~S`y0;y(H=ijA7QWI+-G8q>$TD_CtKu^&Q34Kqa$o;;^>(W|J##{3atO^3L5W-q
zsKc)gY<|vIe9p~z2N(J1B{ER@i6L0n#H*nBqg_a1fd0tWKPd?CyF`o6lqixjGY{Lo
z3X0j|u+Vo3urt*`E#8X(6%IxZUYG@#PW(Q9f-(<MIGUlgrwdX1+C6tg)uTY{TyVtI
z-^GxB^$WAobIB(;UX_G*>bA%wPCs+ciTvM;PMR6NkWmr_N{jrSs6f=)O`4T2SV*DV
zL)4c!5D)hg3xT5S0>9En^r8K~K;}I_%%?2gK{-4!dRbII=OK<QG4RUhnsTbF!By5}
z>9f^bP<i1R3W|yDM{X${4JqAPMIn3B$UN&6?vvkIgj!qAep%<{O#bAVEflEsyY8eg
z2YP1l|6l+lksTT_xV7H(<b{V08EGhA<G!)}Qg_6O!B=|I@<S<~aQm%qTyLy=E>Dr^
zlWy^Y8?kAq6;be$%32bnEB<AJbSB6-8X$#t-7iw}QUAT^L|-w}9cAu!kykEWQuEr5
z?K>|^i`zQ~zHst`GZd1xJ2uVqDAE*<zz%b0Geas{W$t!U8Nh8_Tu#VOvr|yHedwpc
z(KsxzE1RM1f^gP{O*%P#=?{|BHlztmmz~&|mV?f@$6hT2f&(5BU|d$G5{^So%u!bf
z%~7SMkVDD$2zlY)#4RC2!6lHI?baDY1D$%S<RO~-(051T^GGDp;mdUv{n6p#dpa~3
z36rM-?@?)hyGQwf6Q}UNy5aLR*>}^IqkJ>lSA*f4n;}x9GUD9IU8KXlux-e|S~}a`
zQ^w{Pe4_xvK3^!j!sm~Ie;elIpJS_WiDjLS!a>ddPvLIl7X{hEaa~0CV4o>?=aT63
z^M%PHyPBuRkzSB2BZ{9o)1%mWw(NP(oBP40C2UYY(SnUW8!7*e5JD!0uK%1z-zE)7
zcVHkGjZnrPtpzmLEJWAav)c2<ekc1$b@ox?RchY8OCJ>Tr-in;N2sC3nO7@a)0=l`
zNT_Wx>fg#H5RGRw8#;_DUscgQkhpng^yB_oN;7&^kqN}TQ+nroj8{dW)@t%pf$PO*
zOMSeuOt?KfxSeJWxLcDX$~62Aji|x3>D`6RoqoyKiF<(fa$Wdd$p+3W!@m?j<TC`y
zP%zVL1-MfrKU`gNQbfs%fS(_VGUt*{=KXa=6!<0H^qRb1-z?(I)3v4vb|G0Tia#cM
z+6_UjJ`U>Dvw2bn!Arm?fAA3Tr>gV!Cvs&7f<DiKa<*!HOCgJo%Yfe`r2Gg+6ixiQ
zNA~+GC4}@5KfOwB{dq2897QxheQiS)u@TjNEcQnGyr-zWdX~T*3Y>1)APUw@&I@n<
z+0KE__N@lfzZt|p&3k(l)KJt+HBX-=2si?q>H9kA#$jG&I4#CRi@(B4WCDp)4Vu7J
z6LPm)*g+A$3HYB!pP)T{jvLOUGNJh6A0Zt$EgO<v{;~MV*kwlPDGG9Tf=?q#xe|Z&
zY<kfO>#lUN!-pD~P(&!GLqc+}Q`wp4Y&m1xhBICO2DO(4E`~Ij&7;jXi3QQGa4&vH
zpoly;77d^~e-i9l_-lwYf4v-VU6DG07V1;|gGI>P*ll&8h$(bhn8uBK;Wk3iv4GWJ
z%u#lR*;5j%E2(vFCCcvg7)XKzQJ@-ajV7ktt~Ty;G0N0ZO8-P%2oM;JVGNR>DS}AY
zxqBk&1r14_)<C@r4b6HTuZW)`f26#VO<Z1{B&2eSkz^2ne32N7Lg0R^Ulu)kqM={L
zDocCQ*E@%2%g|N`MeGeA->g(`%a$D#g}QVR^!$JcC!_(Q3xZ54^b4~@*_Y)+<c$bE
zHVAFcy#Z4oLERofL-aMVl;y6PzpfWZ?Zgyu+h+MO)nQX;Ih-DQSct+mR7Mj;zlgL3
zhuhHLTOxWkR*d*TROZ(f`=JJX-45MT&(jd|l<m4$NYM3u!LOJza?XMgM4H;D?ztKl
zMVZk8wN{qVbNZjjpX<JoF5|+7!*9!@F9;8t2%z2y;pC3Eo0j}&TKx8xh1h0C%1WQ8
zdY_qY`}2kNjWa0z5<njrBWP(T>d!I#e7ZxW`swwW68KaqvRxkw$&N50&Lb24mV2E|
zyj=XO#ZVGO-1eL_2F0RcR>$p$B^h7JhyJ?OTRoa4r9c0^r;dUEoEXJ(bv*zdl)E|$
zZ+}7;ov}VwD~xog`kJW)Iwvm`dD<7E1jg*%oIWmU;fvpVE7DfgQtD&E;7dbrflowS
z8Z@Y^sLJ>zarauM;JiV<1M|?zRcRW;d|g<RY>VC~_TK$zg)(guiv3_TochIcB{N3e
z;q_^GK$0LT$Mv?DE3rcCFVgOPPp%lHhRihNkVvm1KtSQ~06ubWwDQ40A;$9W(oh{V
z#B-3uqRU34zQ7)9pE~|pV2g4)2RP$k*ktL;FF8E@Zg4PR->7ysMh<X^kKR_P;n^<z
z@|o2^`)%CXo@VNkS&E|$!|=59Y<TOpw1=12pKP61Do-{BX-*z#u6m<lBYC=rDZ~{7
z5)a9;qM(Qys43q*#;q$RZ>2w;NInP6XrlatQr#mvrQqF^eaoz{zN~W3<`TqVKHiE3
zNz$zIbNFt4djiRG@p#pBvs?lEIfM+EA3ut@@G`SikwAr9btOum5f0Cufi^?s-THxs
zY~_g->tz&4icmmt5bX7%wBYK%t=+0-12hJpk9Gx)XNbZ_2DMLJ`a5mnT2~O2vfq;-
zw!(z9Pl47`PFEVpNZ*1_{Nawa0c76$+cHn^R4_rmH{WDYw9V9~g0{vDGyU-!XIMq1
z)Q1!u@wA?_tLoJ4rdS}HhFraIVCVUE*<Uec#+M)L9?XOZj$5t2!+o3D5mm7J(#Y(n
zH{X=jIC<OB@WPzfxCVWVC_gxHv~5!3P+uNcoTEp6dxO+n{4QC2rpT{ATq}Nc_fG58
z4k?QYX60)o!bpC*@yuh$nw0nIYY}V43ZuD*BMIoE5mTYx-=6se@4Lr~#BKy-n?fDb
zA=h*whbG5btR+$SiUy<2Z|=y|`oGl_+c}i*9uMF+$xUVrTh7)I$A_$CmYBL;eQkx}
zPrjzIb{Z+uh$sOFqQ)?+1Fc8xAFIii5otY^RklqupE-ua9fZJOi-900R9pPc!3<Gb
z_oAlADBM*IfNBf$SZpWf3{b0SJ#a2z5$aBk86aI@-hRrB>=Xn(dCODc)Fe8C>j2Iq
zC6)05*%t%KO#4k@_`h4=i9j;F2-PSZf*S)#pyY+XMGcb+2!&SGxKc_KKcd|d3%KUD
zwKB-*G=$yQTwxP*$}K$>lVMuu5t$mT%!mgo46DI3NtOISGvT`=_2C>2x^184GHEh?
z#X?fTndNO#{D`TbDFb=Xjhb|#L<WuGN4HyJfdI}v0R9p1)5W5;##fnxRCE95EWjfs
z*Uc$UWUnChx*!5`Tg#Pj1Ha<@w=nOa)rcb%2LG=uV{BsPNrblKVt*^9(x&vC($EKE
zb`8B9QBVo8tb~+|r8To6e6QVjzy<&Ci7*=qw4ZSy2|ZkUw2ez)?8-ErrMxvEz`cL~
zDQ^@}lho)>A>kOI-4+W@;n3Q&q-Q`w@jb89-w0@y0{Us>rZ-HBZ~JMC=C7P?R>Sj2
zs(!^l<gdgfGy;1L2kMY78h?KHJyO^}lynpaQf7)xXO>a?W|mJ8!yE(o-0UL}Nq?zO
z{PgW`*8Y@JBFF~;wstekPe;XdY$&Uv8TSOB=Ur-B3@9SIcV!-OicE(K(*QFylr`7T
zQ;88GPHzf=_U1I{FWpH=`GNuQ6*qCf`__tSTp<F^DYQFafyzSxyg686b(nN~_jyA%
zxIM4i_*ph?1cjJ~^AEXj<Nv_Wug^h-$WoV^8-5Ar3zKQEtS_(ke_ba#CGAO6z=!+(
zwaZ$As_8%x7lv<EqLgdROS+A-#BP^dk*PXK-rDCU*(QyCD;r|`AF%1eP}z<Bo1&nh
ziP2ewwpSA*#Q@`o<)G3k7Rk{{1u6U3<qogh%u8rE^OE$QAGq?;UHU<e0pEI%X<=iY
zBU94}$w-I83sV7({;C#k@_8Sq(+9>UI8cZ-NR|_Yry`@v@`3#KxS2UhcM1(23{*m%
zyTX1U44tYd+TT4Qb({!p5H>?W0lbZDgDAaaByn;kS2vZlKQfvosj6IigdR<-MaoMy
zAn&EXLx4M}fgjvH9}DL(7OvShr7RKmpG3Pe<@dL|x*>m!q2?I5+T}ykjm<`8fPjxR
zaiH)F9M9Xx^eDUy#6P-%t<FTVSEpCf4Cx~ZX7AW|D;=Blyvz-si^TcEDsvzN>3|kG
z#0mFVv`Rmh-muH2RL+`b_DBuUi8JX%{}Wnw)7NaC-0ZIFCu87^Uug`4G)AYL15S)@
zjmGY@7+3DR|Di7!-?aQAZSjF0vNyKiCh3jR^(&~}V1}e-e!z`kgM6e+H+^nVe&4&d
zTGVd-N>NXFY>)WHzb+2wtD}6U7(D(U4tqk%Vd25~rUKq89kP0hG>KarfGOg46vtc-
z_k@!uSZ1A~yg8(F{W&9?mib4S6GQrPS-fg9X~$$i_qP;~RXS|n;~|T$&!I^<^kow<
ze;NMVf1r3A0yB6kIwU`p$m)ZW!2eBuLUj25#nV>@MAbcipIvrgSx`De6qIgB1&O5_
z>5x>qQ9&9O5EKCkQBZ0{=@3EbSd|6|r9+SsP`Y#9i_iD>{(sLsb7tnAJMozr?Hqg6
zoMB00nbSEFAqd^S-qbep67Dq_F_=KThA;JXnkD|<Hup`HV3u{I33)Y3e@GY|G-1^O
z5KNdw2UkJ9fezF9W8p@%5LqTTR90+Ut$5J{GLUin#IS_7(V(40G_x2*#vx*?YW(+h
zixEH1Nfz_!^|4Er(=OeHndZ@jvr!w+3l5j&Kdg?sh|<GI>e5F5ZkxYwslfNe6>02u
z=EVI9XM7he8eC3={>~NC0pE-dq2jRWc&Nfo1V>H`)5RqM)Cs2Lox>d!86?i(j;C{}
zG7m7h$)rP+_F{l(&xz;%zNB{*c=xLh%SnsgYE42Qm9ddGhVqr`GwfxdkM42sPt3O*
z&I)QrQgHwK?#?km*uYXT|7<CJ;axvQQ~qI_rb1HW^q3c$U_D9R<!WWuH~(p|Q=cKI
zreJ#2&8~H_*`Y}klf%S3X!w;d|6X$8X_a@Qk6<2+#~YJr!asd8OyI%9>|f@6(a)6g
zhiolmkdBz`qLzic=IM1BtZ`H(^Nxq|!yAuWHlDU{{ms0P+|LSlySYFVpW%CWnXIw+
zEtAJon4HH2aapCoS7h}1TPjyOc<&roMAk=%7xmMe#^Q=0rpQa2<6rHr_j|P^<a^HO
zJrMrrXKk{*oF!;=<U$OjT?)D~D-wx)4ZydjJAOy8cd5pD>e1Lq(bJ%R%dc438e&UD
zQ+_;Zxt{)7`sY+K1)~r;{{vBb0>Rn{uoHc=8y0%>K4Ub}-JI|txDP@~LRd52{Jo*;
z-S>kP0?Lwf;GJ+?tHN=8Fc#a-*fMeV?hby;)Wo~rZYtn@3bVtHAUq15Kx}^<g++fm
zL6ZAMYb(zvc;7O;V`xpdqk7)Vju>1en}2VwD=*_E_Hfi`cGi1cO{IRRA|+i<?0&7A
zC!^c(J2{#&$fcpv+BLxdjS}7@`s@cp^9bDGe3tsJrQ2isivcMEms6^VBJ}HLsN0Ok
z<g7L__J+mz9KwF!iqFGz@94v>-Jxk^&*-t~NU3t!bwp#d9bUz?(hf2ipBAzY8a%F&
zyu-i}#4I;|<%TPvUXFyDND+uFQi)_UsqRsN!D~0K08Syo;IW(`fa-HrqGWD6UkQd1
zd<bqb1?WL)k)*D{Btq9`siz!wqWu*w?mFuG>AN*=+e-NyCBdh^H!YKSZnVo0otXt<
zi&p^p^nH**jc`%dwE1DOJJD`+V<vo^N}H%S<w)gYv1#nJ;HEFHzkJ`AUu=p)du|R)
zT~026I3Jrqq^3H;!#h@i&fA_tV~Vd;W`~AR{7%N&^6J$swxfi2KnNzW0ro#b%A0cM
zwk@o^xu1tu-{FOv?XW%B!(*Y^i-Oc~Dr7q<$%C|&S}Ra)E_re~Iux#(8xA$aF$Gdv
zO%Bmb-Vbn^Wl@(1G2@r*${S>eVqC~=Zv_jAR5yA#;pvidSb+xQfxQr0EYfuJx6Tqd
zeNIn`y+q*qIc*7Ly&J23s(`;TI|3>4il7wt7Xj0~n+)u_^yPG!&U>?rbKc`kJ2y{=
zccRx%H#4x^${w4sN;Ag5P-lie_npHXtA{W3OU51>{({{;Ju0QchMi}fNorGuO-325
zwGvRs+HesX1m>LlTy!o$zm4(A>!*c&90qhX9JL0v<-d+Y8&Y#Gyk;0+tePcI#KHv6
zJh{Gz6Ri;xSi3N=+M8`YR;D$V)BQl^9`VoAjizX*VjK+!kc9Z;!S*6AI9Amy0N=N{
zUa{;x=~k2uQ~i4~iNwcNd&-wUcP5=K7IlV6aM|{TiCaI^l@Ghw-}Wer2HUaUR0VE$
zYMxw7_#>e76wtCF&XVCwCkWivpAU@ujSJ&ZS>!E0<@hBo5XQBCY)yI}Nu+GP4xA;{
zh6;fHd8jy5E{24lqVmFJ_E*xRUc$eA93e{v99szva2qj1dnmQNf!uFGW|rk|&pT*J
zEae%%Az;1fS>aOsxLoiqkFICi9|9d^<J&Ab$1XyrGN>UdXs05(%S!x*KH&2Sf5`27
z@QJ~bEr9rsxXbQ&_QSsR+A4MSBwvD#%E9lUl`NB9T{}^O(<~QM0&m0)&(~a=kmAJd
zp?r0j|M}54RaaQq$&8gmOd_aW5%;y`Qn?WodoZN|3Lk^kDOE8gu|*r>{M8T|7@RV^
zoC<-7Z-v_^cDmnRe_&-B;Xn%pA1WNL;eJ2ucbI$GdpTbz+T#{f(#3do9G8!A@a4Mq
zx9VyIDSoBHaqV}+BFB@QobB%}4vO~4Ay|pexJn<c`bgt;fV*K6!8aYIOAdaP(AjjO
z@SDh*3&RUcuikG<LdjjqGw=fjIf~0-jI2AEGE^TcOZO1Q)dv}}FFt~^iiPpWNn9O&
zy%+O)IuwOLKmi3nr*up082R^4-UpQ9eIE;lKU&2>1ADWY@M-obw1+s!-|}mP$)#iY
z0}3C^eEgnX0QdI~zMmm8`Y^48koayHv!i_0o}8j0f5BXm#ZB6Uk>Gfy85azjsus;}
zi}HGUJM&-jQ^P#t5>5y_;R^Rf=vGLOX8x(b>Ve2K2+j}c7!zjbB2p40*8CBg@OR(u
zeW0eD^IY_rcF4KHS*Hi5>!NCqY61JJC@gWfqvDc;Zpebh^8)9}wu>h|D9rkSeGC3b
z!gz$fkLespx_*^0FFK~=3+3sp6?-v#)Hg`cFt1QhA;n%tKC0|5W6uw<i!FaY#5~q0
z&kZzBq2=18pwpSGjrINN-=UN<WE%_m0?n%TvFZ<0AYq=tDKJD5gg-~6ybr#5U3TY8
zBABpcwwoTVt3j;+>Y!lsIPF^m7Nzr6f6|2L{|-2P-&%WZg&PP=YIj{{+qsp!clW0%
z1gyvie$jo};o1guJNloZ5Dl^E9ZbZI|M+%;f(pf6wyi$?=>gsZHj?MiqyGAeU0%k<
zErxamN8{2u-v3i#crC1w_7W23Y$;BJJ$8H)X#i51CLUKQRy?<O&XfM>ZDgoxIFEG%
zoQvrH7&qTGIFm?~^5*@=H^8v8wL=T8mWQ3qRTaX|23=Yza8@=e7hiVV{EMdJ#*1R1
z@Ea!b3(=h|yf-a2%KXWYY=k;&nlFKDt^L=^y%alJuN&EgT?OkG9UdrvOxPJ{M9R)O
zCH8WPBNet*7%zFVC>@N*>QPorXe+t4Y5LVFliwfSt^0v|ss5HP;06v{<|cyva%wUZ
zm38^e-9{Qee<0b{MTJuIJePH0`vQ#~4c-!N^nMJl)Dp6|^C#Zq*KY=0Ht4{vV5QFj
zkojJ5IM*`Gk;4J9B-J=7H`yy5uUWeeL!9rp<ozPpeVk2(CnSKBo6eZtV9rm(;CL+z
z3PWO$42kY_m-+3-xQX67e?#IHl=)fP+!=L61z#r+$o9IknmBK}dtHg8b`W8JdPGnp
zhd~E(3%e+D8QoYBsnzRW$?=Wqx0#lavrE&Bwrv8}Cx00!x~#~!=!=(c8LKkbdvRz5
z4w(#Gm=0w*V~WOhAT#25hTW!wr|(w;-p<grakr43G(*G5l<~)aI*>IyvEye2vfO0y
z;G=09Sjmd+gn_`W%O2vcE+GLXk|eB0`(Mo5fBm&Hd>PBD#{4i6p)yEAcgvw#`hUv?
zyvB!h7GA^q$bsEXJM~(*;T=cyu4?V{xYp)z2p~cC)PQ4Uj?@SSpN@dUiU-<9a^pnF
zYLghy?5#&be8>c6b?@^ShJCVLuwyd&(frANFF_7=MiF`6Llo=X^=`mwo#6C2>_)*7
zGgkQ`WRtg=GZ@7ZX1q{}dLmCeL;_p0{;2?GDxP$1@#9gVpB%zny!@T&3SF+d2+6gK
z@w@o|75B@N)qCRCY_resIm<NuQi}f(gb-L#nkLG=s~#Tuqkh-{G2Ou6lBJwuA2aG9
z2x158&hbqJCdoB!+kL+I&ZXyNzm2NpkG$hm96y8&Q@BZbjS9P{-oWW9LdyKkKsNRp
z!bT9Odv6`zS#jL>4B@RTBq`KjOva5RfdEE;@8b;3X|ChkR(Xw+P^3y5<e6STVFMml
zd_Sn!Z`!z9FFTL0U+vPLs~CSNTslJ5PGrxD!d_sid2!&TncH2he?r`az=#P2AAdJ9
z`d(%?+3b(P`b&O_w0=y>2saHcr$FGT&xv8CJ#V4FA5E4oW*8%p!HSViBJU$fV4wMN
zo?T=393zct^Cxq;+3Unmz#c1H%);67$kqEJBL!QLO2QbFjqt~2h7n2xvuk*^(KdZz
z+;~{0G$pWQl+(_cKA9Qnjg{|^v~BZv!Tah9%XlYDEQ2h66q9Fi<$x)5sFf1O!@Y$8
zZv8I^lJO*49j?NpLBUziWn)K$>~g;9-Y^jzRltV9``XT@C-l*qTSgFW>F$PskuST>
zygmfIdsyZ7`Yvn^Z-hmUo34k)I*V67>G2DInYw)U{;iP{1=7#ZP}id7)9gQR--MMh
z!RD;D-Bdsd<&b8A>%1~K_F+SXGS1G2O5#P137z5K$Q(}-*IAC?Z~t1ugs!U=bqbxP
zi-rr5$jCDvRUDJ51(RAR7yR<BGQ8@RjEc+VlPQ|yU{<iZ$dKW<zpn9ABWGT#;c483
zhQC~Yc6BEDkYYU~1Eb^xRmb{e<;IZt_rJG$KHq)zlT)Q3{c5_O0joqs7kz?`1gW$e
zqHhu%HaRSSHVY(nrlUrCVIeW$V6k5+<|!gy4?fbvtscMs7%SkXYTHvD(ec4t<RYQz
z8Bo#?7+F#>GcU?J4#9O3!JJ^vYb2dxe=qP3>ZYxqga$g2iAKjvn?IjaYlLg55WXU8
zV$Y+K54+!%7X69{5z<`<17h-d*DGwqf?Zl?SvsuhzYKW{yk`P_OAI&rZ#llBF!U+V
z@AypWbMCo(OLG-7;2$h)sE7dA(|-~X3;V3IkgfFVXus#=5Sf}jmv$sUvun-mu!E$!
zwRqJgQFxSYec-MXO!w1bCA@4%U|EzB>vAT`kC^w1^<Qsh2BiA+rs{()mX8LWdk?F6
zn+W|je=e8#%gud*7H;#)saP<(@KD9M!6GB;xIq2_3f_mXR0aMoO-k@WCWlnwdMnut
zH=HNy#w)K`5~B*^=DWV^A<%t)#NS?!iUq9yUE)k{Hv5zIJvu)NM+(fF`p_i}Tf8S*
zAW9Fs8LW}Frz+i+U}5^2S%}WXx%=fN+40DU)8z8UNPZ8D5fK7UWPb&P;Tw=KcZ1nP
z>N2E^qQ(a9em<R98U3m_Ii8jW#w9aij+v!z5NvGDTc5ulwOL-$|K}<3ym^sYleNuz
zDfc&&SQ!vk^Uy|&v>Vx}M2R)dzl&N3c{zc3D*fT!eb4ww)!Hv(S9-rrRt$dP!5>v>
zsnPa(3p_Me4It+Uk6DjP&A7SlFxU|7Ie+Y1ODunXZIv(Gu23rO3Jn&SMgNyTyA{hB
zR&!=T^pVq1F!(MQQ!~-z&?cl1YKtFamQ7cdgkFXaTGe4ctFC;GxVN()ttMi{LDld*
zkDsYFHYly8P#aCXGRFnE_k+*gpjm0Kw-~NjS*kq88{)JOP*v4R*PG-QX7*7+Po$d?
ziwgsgUUmJKCkU1Pk0dmW$qRxuIpYIAhS>N2v8)%fV__bYHx6Hml1i6<dIhhBOc>B!
zCvde+`^;_0T{^q8nH2WMIv;eXk?lyg*5}DSyyhh`aQ6jPoYE-^wf=J^xrcr<bzWgF
zor!+}qX1j4NZ-uKf0&LiB*RD!S3eF_ILnZUDaXkBpCdzC4CN_4GS5OjXU;6B^0_)K
zkPI{<ax=`Hy@)koWZ4w9bf=1Ej)`m;b?dVN(}>M~Dnttfsbnq|y0-c3;*#G|lkGf4
z(PFOeU%53WdL!C1Z+C3o7R3fLJ;&eU2u}n~3qbZ7@w)sd3A~S*$y$4<e)Y~E4Y}M~
zsyIsY`oh(sr=8EUZm`8mS=N1tfC7A;io~!le<CUJ>CgVn0>9=ez$j)LZ4!a)eny_s
zR4rMkON;gGZ|Ak`^QM2b_J=J=Jt?iCy$Wgybx+IiOVVsv*$r25JCXCSd}8Qs$GEDo
z{7L_s{l$`x4hzqsD>uZ>5UK<)YPN&k8s<IYsX<5axgtxHD@j)3BtfrD6YJ2IQ~(t3
zmp6m-fpK%#Lqzptw!;M5r6g<))5;Cwf%%QIDa}7DF@tj{uG(kLm%bb_c@_F?uBbQ{
z$c)7UJOOr=`D0StHp~e>;J>V0c`M?+fLD%UU#20IGtATkz4q`0?JG4m9ly-+VpaCU
z+$7BlVK2lv^80NO*Va3-D9#Lvk}GC7^?F$sljn){MM#XUDu>f)6EF6&wAYW7pr_B_
z2Y<WC(W;gfN^>qhUQl1x9Phtx@1g?OSdJkeK14Zld0!x5D*kv^OoW}7p+jYkB2jYO
zAN$M<8*d3;Rt4vzFNZ{jl&l~OtyGicDyt%+L|-nJ`jB1iMjbvTQ$wA7Xo3l(U_({B
zT2tA8ihqL1C6OBIj+)+;$EtzFuk_!(z3?-Qle5;-%jto6i&uSiqY+XF)ou^2!69)8
z7mr!aENdK@=>zQPGutDU*dS@I8txak{dc!lqF>Qscbu*0aCbTI^u<5I2^|I2dbaaA
ztieMZH15vYdRA}9#2;8)FilJk)PN9|ME|*kmUzFd^O>oyzgbexs}f#*SMMo9FI+7E
zo^|ji9Y@RYo}UTZe`~yQ4Xg@0=sv%cw1<pdtAg=y{{#DoSArb`BoYuFpE)**5u0S&
zJ5L6;J$1vZ0D7%_1&L=tpt___oxU936Of6yVSI_(Awl%Ec}3Hb{IFpt=}^->TmK}u
zE+dRQWYLib^4Bn2?6oe{^dEU{pmp8U0+H<~e7TIp*vnTU@vffE&d(*iw228nKzO>O
zVv|5!vwF+~`h5EKsYNx}&v^<McVn^*fMC^X5NuxUo6<wvX-Q#u5$3vTcTcxsbSm#d
zQy14idT`e9PmArT;#;!k`&iiz+=I&N*SW0b$@WCQ5&V%@pmY`pSVT(Ia}IBm=ox~c
z=8WZJpnUhXEIn%$vag=KYxx?F((j+&o$i`}<OeX=ow~@M511R{;Yly-MI8gS7P&&C
z&L9%oaB*eRkW_xXCmO{+_p2XMIK9ca&J}WV&|m$L5jl^t`gi8j&>z8k(;6enr*g^R
zCYMzqJ!d!zZ)}HOr!u3e(u=fUPxzeM%PE*JF<_^q^CfAVX@Zx$(m#rW=0_{<b;`%9
zG;TV3TK<_FQ~!V0FP|Gd*kZ!Qg{)j&J}Nv?k8tCJ;~h9ZM0^n!uQBR>2~Sv1341aB
z7Bm1|q3!wa`fOwUw~oDW#n7oAIk}spRA-@{?Ykc2SR3bzZg84c#=f?*9~TogsaF>6
zJaFF}Fu#h|va_#w<BFI*^vKqW<CtAk{qMxySUKdY5Hip6uZ#E?xzbHbnpiKK@TmYF
z_wL)Be;@Le^7^<`8Bj(qy>nRQ1tt=X-&hh#-}c9!c;h^v>ML=J_0_4YU5bRpi*Yg6
zyIu2~^zVL0VI6d~OkXsFu1CJR=>Iu-xjQLq!CQ4a1%wLMqOk_FD!rFtL$)1>d@7(1
zoT~9^?^R&*gA%s&+u>zoi$qS2z+J2UVv1Xcq#t_*WO$7ursd((S{)$kpEY<dlGH|y
zrvF)g{HfbN3WTpZc41&oWC-FhCnxvtr=BM$Tmf)Hc!+ic@U(eov}u8DZ#iQ?FKeyE
z-*vxb(J@p)_c|H8kAB0uXimOdrQ1wg`M_g-rD>31*b2-g;+8x~AKmn0{AtREETaX9
zB~WrvgJ5{!ZPv*WW|AMOQH-@eW^`wIuphx2<0&KVPdvT3m8`D_@}C-{3(k6?u;Te?
zNSM>pg8il!fO*c^q!C+Y>{*%I-QqsyW|*E%tE<TJV&~WO(PV-v(k3RNXx<xzq5E%*
z2_^Qy^T=A(E@_xDe7gCgx5SXx5(4hk^QG7}{ah}bRd^@zcf5=mD^QgGM+*k*=gb{&
zp`6VWRgCJKG@ryie;7kWUBzadef6vT?EoC`t2;Ro<&Z%F1H8$<gVVmh0wOIJQX}UM
zrfzv9lm=Kkd8FNN)?r9JHwVWYR>mSR?|^F&e3P))O?aDid_jH^Dws3q<^bO!bon~-
zg6Vs>SX8|d{j!+uyd6t+wV=(x3<<JE)~$Sh7qU$OCAKHoQ5el`I>dIrJnSPKKl4M5
zOTW{9FOA9&`akA5h3o(1chIQltu091P3m|};A^6(7wdR7O_rxBxA#1}mWL9X>n9HD
zGcK$9V0#+R&LO<hRr2veTYGyg8pc}MWdU0G@WLMwD7ahDWwiN$aERAS2pbKqzJsTs
zzX&%t{8I(M>s)o}uCD7mk)3x|;Y@ts=*v&{NY4en&o~16e@jLYHmznLRK`8=-eSD`
zrpusZ-L3(t3!}1x_6<b-zJB-oXN?kxiyO&P$N?G1!bKu*2@_O;nQ<PA*#qYHjJ0o#
zaDC!`mjYt-LDEpiyu&2Rx_MYF2Sh!TvN6cB9e$3a;u?V+F+o@&WPXD}9Gu<hM2Z-k
zFkmLBAcodw0<{}6%iR>VCkHKF@ZAHXihGjSJNk#?FC-|jRm*x<hxjVm_tsCt0qy@*
zqyshX*npH_%5$1vo8hMc5aWO<bN~A0Y9RK}x)TGVElW3LN2(A$a*)X&JShbR`Eu*e
z`;>3-MSWS;SaN-te<4q63N~4?#1GPl^Urk}0v)k<vhTirnV#pDA3`44UD;E@A|<*r
z;L5Ea(|cTu&i13^J~1Eni^y*{GhgcH6#Vr%T6Rny@EYNTD+3(*Yw|d7p~U%VFql)c
zB+|x<PIAFZr-zRNmOojfDwkPP$&S0_V|3@$T>rvxlXX;mOL{a|IFd$S5X%l{+v&g<
zi@<Sw@ss6FHFVL?81cRRtboV6Ch|^ArWU1zZxHQi=@i6CLEZfsuU^Zx-GVP0V*u0L
zFFaV1C^Tl26H4C-hPL0KwSP_CJ-WAX-R-s<O!qvTH$(1-C<=x}3}j$|Saa8As1D%W
zTu9=wxH6`1&|>(Gd&RfJ*`(*ikS0v5i%3L=H4%yxv*T6~2;7wgTI?IN4NKXYzizp`
zxDkjXT2y$usbU{pl}F4?iw}qF31P6GMYODPMv4w>aW?gPTa6PE%RRwyDl1a<hnL-s
z9JrVcBwow^fxcUn(E+AU7>Ul>Dg1lCl^M_wnv0CR-11`VPt!a}EX#`M+^x!0+S?UQ
z+TNMUtzx~y%@fxhPb-$nqc{$%wdRx5XBkL*xLCnC;XL4d50(hwmS*17+us-a<bdUs
z*z3HdL!a8>M_;O~+A`?tVvH=?1Om5S1wIC4m`PF;MzOvUSV-!zELho?ly`RNJh0Q#
z(w=$O5iplx?A7>xSOA~mP8W+x?N({eA4A*B*YfMH<eNiseFDML9Nk<@jVcDKPf@H1
zSl99{8>sbad4Zvln^#^yYlOyYB#aR(>->)$x6SS_y|g^PRa|$BW>W0Y98o%8&_Bln
zo6rW+-#khGJqL$ste=>rB2a|Q+`#AVcbMrpJGczc_&83OQc25Ry$ZILh7$WQ^}!Xd
zHI<wZQ&Wwh&zEzgrndcM^Z)9H-``P^PocT3V{eq^{yjM+yg(iFg}Unc!r^`AA8dy1
z8DF|0<C8Pz`s6DIjcB7ckB87Sii*ytSIvkGguU*F8}T8>d(!PU$>m>!PWTwoT+@sd
z1omCv{m=|CMeEHF#Q4i+wAlE13A&5+og4GJn@q^lHHtHdkKKWEwy!V_Dv;*pK$}F8
z&gf3=q>yz^j|wa)e5WqFIozT?c1N$Ou;JLaWonD*<YObqKLY>lqE%8_>)^m0;=zIO
zP&bn{FKhJ9N?$OYgUSd8R9rWtRX=D)A*?-wD)cre!SNJqAQ%fiLhfTx<z2;y5=4l0
zjr4>X&WCdSVYi#q5G_<W(`T(cKBQWq_&gkL-UU07fZC}<)Ix9phtwp3Dk-L6O+7)z
z#K?syRkl7ccPj^D@W~h?ebS|2UTXd=sQ9go2o2kxfc|O}x_dCl@q<KA-8azr+P+4#
zCuGwwx@wXW8T)!^Q{3?O#A?mNYZ-S-O{J0riKQ?_$nufFOhOnD(QY2#TVe+Jmy_l<
z0)E-4wAhtBOZUBCRB$(Zj2D|PKhjwvlphGOfEa~!{$`BEG`->aZbe!;5X<Hvb~~l-
zk%su&!b&s_>K}{|(uF#(^_>^x#baRkTd<C|EHpyxP<dTd5?sl#!3W`<Z)p47pB8_I
zxSH?ZJ0D*_+Th>=-{AS1Luk-3F3C|xOym69%73`1;wh}xq%zR8T>=I{jfs{a)YBLY
zZe!}}3Zvx41kfh%=kx;`StMoju`NC0-zka#ng7^b=5guA2L|{%Ntw?Nk_sL<T*d|0
zz=hnEU5nEiJ`w%QD#fkBpo@aj5c@`3gbH-r<jf-cP-ZV!eXXkZ6z5%Q!<Av-Tt~0<
z7shlF+en?C2A%!M0mA0*`9XU4G@@l)^7hVhh|hkW`S&Mk3>39aY^P>JF#fnO;EnX5
z_;+_Rz*;NwjKRy%0gB34(x3eXU0*kYF0>unz5f14HGYT`foWMDIyZHZ=yVNzcCgoX
zMPQ8lYDVxf&**vruOsD0*3Se8>Pw1OHe!AwTHkQXzGF7(W0)JY-}ctQ?SBUlSkT8_
zB*T8Z*3T~xNQZg0djH&J_2K@lWssthAx5<Ix{x*P8RLm}BjZ0#$J8LETLoFOKl0Bq
zF)t+K@X=Az?yL{L4^2l_SMd=Z_=z!!mu1|go9GF50@n_I1&QlkhjfY_Lhq2BUKk^;
zm=Sqi+JKn0lkNfdhK$eyLg5*}oEBUv+Jcevb>mjy2%Ipd9yWMy-;o+S5h7Q9Be;XN
zjP~~ig7wfgkfW&wW`|}Q=S&mxAd!DGw8*u!^G^Y7$sg1`4oqAj5UAnP`oHy+vTifC
zK7_xh^su5nY^mwnVHq|oAMm+clkm$)*|zNWRTa6E_o%av`L-?PZ({GY$e!3_Fk351
z1l|Dl^A(csz3)8Rr#GQJe$YdG&CgKOal#(z#GHCzAP)MQv{80W@(_;K?&TiN#?md)
zk|di7yfh-8;ZE6RyiySR<oim~v(IlC1bD=4TqB=!`PfL<Zz~=o{k!{RT1oLF4LqCA
zQP>gWfCf}dJx@DpOy@;Tp_^VY5w3IG8)N+ifv42HP6RaxAwpfhn7)|}Pq-wkt9ni{
z-lXqzUi}TI{~${_^_XBOcI54hY!7A3KZ3y)!xGV$OlAqu$;OYiAFefRj;-xKwS9O)
zjF;P;u!xa`YE2UYtr=x`zS$7MaH}ZDBdw+Jh*lux!MZn$Mq7o=7I^^tQPBGYwXGbE
zO@4C@dqIOCV&wS{If+%2Z9j&fDDQ(tVHma-%%d3IXx^OBNF&3W$y*rBa7Z33^G~}4
z5@Iv}`ewwgDUx#Azv`soHe+MfFI)g*H<2BF7ZBnNdSYGp22?v8q}1hdOei=v&i%NY
zS0J)05ae$<oWA=9atz{~r+6^`w(+J1UkzWXDLD=$`ySVL(rIT%heR(R_v_nSiV`$;
zA~~RzZ1nB0q%3B`_fUHhd6RPc$C1955$wZ`Ms$m&!g9}~QeES*f{rvmmCQX5#(0tp
z^VhW`h4&$YkvREIDL;PqBCtAu?UGz3F$e5!9n|_fC&v5O$|bC8X(cr{&AMc05)QmS
z7oqW;O5^mj3-APr8+joXBJAjdd-FkX3`5`c1H2c|rYPi-2Zy*kIBMBt88OCbzbBRc
z3C!iuE*{vFg7)h8eq#feO>j|EtUq<7!uRR^ct&0LAk-{7xX%bc+MXJ&PxE1U^Fa=h
zrg@;hIc%jY_DJ8|t9G+Ia+8dG7dUePM$k=B$ac0%8c@r72;qol(G_e@e=I#ee@@#G
z0-v}imPYAq4fn9k=Sx0yC`-F}NJ{h+Cj}vpDWsGqxC@s`fWr<QG3}i5E;BavI4N>K
z*JF?$7$JTVteFi}A@?=xXvCU0P0G^n)opQ}Pt8Mx`$5jzOByh*fs4T>`ZM149YwP*
zaP4p1mKRAN+}bkavCE>4DXH`+YG~;=#PzUAzOC8Y$NcMq)%vg}?Wf-S2Xm?ncRp9;
z!uxcgwuZi75K65FC$P!z3FNm=8UeRf=6N3byR)RFWQ(>0<Fzr<aJ~2lp2O>J#&5$=
zws*!4u&m&JtMKggybi^TV4Dsb-$}Nk(R;JAOErlNtC(OyRZYFgNxm)Jmeldc?D;Tf
zcbqr?gYm2&1HW)nM)YShk)2^4*=eFZ@Aa!snI}mQ8^taPZS~UXopZwX2l=tjIZPm=
zF@^3V6f7dm9tkpI&3lUOxT@8!3?AihBW#W`t_HejeK>ciJpG`RjH>h7y(c1Da$fwm
z5T{Kf`Bsnzj`)>U=!7q8NsDoLqA5I#-DD`_?b<u->Xx^#>x}5FB5=%FX`2BZ_UH31
zf?FR$=gF<68W&sC7s9RIvIt_4g$(i_Kwix+cJ3#AJMmw4KCrRSy-O!`=fVX?ZHs(0
zqwjL%w6725FZG&O&Mk>@Joq&-W)Po6I%z6<=trz^E~E?y=IcO&5cB-HC%bHa@bl-d
z2T;RS@5&RTxv`2%<@Kk}-zYrZM{aRhn!$CM6zND$+_ZRd%w;(%eFkz?{s{SI?zQs$
zD!3O<%L*TiW(AY1T=n5xUncH(a>Do-o-W7)Qgln!OHL@<=sERa2`-?5Lyf-{(9om@
znhYI-!kYHwz#}xgEJ~E4>Gxu|!DGW>Uf5w1#Xq7*&NY0eiSOAuP5t5^h&6dMy)v0G
z5Fle@_DXe>SNB71i@Ht9NzG-xVGCMNwfCaWmF3`JWz#FeMlG7Zn|<rfux@#ut58p|
zMaHPt)yaoV_H++<v1!HlZs{4p9<@Iap6~Y@FQywpr5q?DsA>BJ)#GLPt<__;;!AF)
ze{oV!=#6DyEj2Oa&85Bf;@&(sx7d?!6Ym-yihg(gtse)<N(<OQb6_ZKBMoyHTw3xU
z&7=a~{Aj%Xc-l+Wch)B3pHka5rVhI*Z3NYFXd|$`qslPus}qY(+q1Bb$TNOf-f5l?
zKw{_-9p=o+96Y4hcXT;hFGz<{>|Vj=-ePASd|;&=t22W7>Fa%Z%R+(U`M1l{izv*l
z<{u@+{8wf=t?=I>ypCLgfoYqLTgB<AxxG<8tr6It-XHs1%nu&}>6`!7pf4_XbP`xd
zO4g`SI_&Bcd3C<Tz&?@PE4idao9%83^LnHD46S|jhU~ks<3xF=aM#a6Hv`gax;;v4
z#`hBrlTqV#Iw(A&PI}zzL-gpuGqD?##R+4*Zue41pY6S>^m;eUjG1`S-%HX*ZO8>(
zf05txix%Mc!d?!7gUaeQ6C7&Y#T&~RWNZD>zPPKvNhLHU$vQr=$<(9-w-3jGUh+aH
z=zBP`gv8chz8Qn%r=8D~oL&lw_}u|siP)}p;CJ?gSV070`qe+VT2JVMMf9bEh|$V4
z6vpLwrpWVE0_WSX>?=w^FCX-`7`MZaes}+!fdOi7aknh5i=ADQpY!DbmUn?h#Q%An
zw@B)?yYEgi63cX}th9E590Hu*A}^#kqaqWY@Wh8agk9oK$BPtA!G>1v6*gkvnr)k%
zsxy83F{L8ib>X8TiW|QbzN|=zZG<Lu<R8Jc3sBm>KIUn#uo(&ilw*LiT*CD`L6+#h
zj8D-pz>YhlMBs#=XxOk?Z>>st;`2zvYlU8-SUvTNuxhQ%?>A-hG^E)yE3}2G6wh<x
zfBg6A>Jrd=^fT>6uJ=T;h`_y9S(h2vK(SWG6%Yeq+Oa4@7KqzXNPh|{EY?z+jIWKc
zLo>MkAoTW|ELR<BemQi$S35_~Ia=_4Z|u3ox+Z)qM6qM{;stN1F$C^z4|sJy;=&Iq
zAW5T3iN$b6uMOFtAPnx#Jru?>wNb_wx%SD>cxZEv5ovmTv!nXS4I->zx~}j#S9j*v
zx$V@iBQ2fExqJyVj8ND5dYqIb0(ad*2<-k~d{_ZUVQNt%0wLLRs4HKS##svNttLKv
zJ#A>+cbd>mq0|b5;T<n$F&J_|okCQ_!_LzHN0Z=>QMgnk3%7fS>1WO}17hAu>=dQj
zA&KgKZO^Y@)Q|kk6JIh~688~ABbMhWv8&f&aZ9lSyKS$tv%{lnmpcW>@V{sM8T4Lr
z5{1fcpoM6#npZ!!+$O^>J|=QP{gd!pW0({fR&H&vkO8D4IYLWb488o*37<ZAGDG3v
zAh$^8xjrrU^MnaDZGB)%{*_x7pzzRckgQ?VFTlRFLyFt3F1Pqrj1;TTNKG-Y_>HQZ
z41CFO_;-(aikYgff>}Mvr;+X%2{O8L=qv>b={JcyAfC2@mz|na4pLMY9y0h~^sbqM
zXN#N?+MGi>nT@Os0QFw_z2RL9$h=@k5)NXqpCzOw`*Ld?;9jemOd>&sGyY@#w#0(k
z^g1~tl!C{F<S!KbKDC~GuXlAXq2SkucgTmy6cIQLSb7z?9|K@$J@k7^<Zs%rZZ7m|
zRFp15Kj2p?0#xG;DG)Z~tl&Un;8SKGKpNB+ooD`0A5D6f43@C5;^SVn9iY7PVqEm_
zGt}5_S&kAbj0No>UFc`D)3o8Y#-ENEKB8=Ud$*miq>V)|@D~eDrT6zD`zQ_f!a!_T
zh{C5mUhEQbI#xIP=ei=&)Kc~pO0Si_qyRX{5(~M1zzM%)A;WRxuM)u+ktGOkYhnk+
zqglGwa`@sO{_70<><AmP13Dc_i-jorM*?zL>hJ$hQ39x)6O<;8dJ|A!AmmMjb*j5~
zhetS`3o-p;M+sl6d@96G$sI?LPwx3UxvUxMm=;Ora8d5?6}TO~EK7;KQ{9EaJqHLh
zzEDK{mHH{P?l}X>P@8~eNZ4$csx!S8xAZbLgdLlUtA)73YRDF%oT=eW4PL?d1qlO^
zufR2vZ~@KW7-rr)%MUBrF?7SJZXJuiVNPUcWT_~K{^sh{f6QqBTH^6#jdOhT*trS3
zgl~F|s_*Ib6s^wMA1FW!4LkVn>5XkURZ86h8gLMhQ38T^(U`IW$CoOR@66q;TdN<9
zzU;q6)W>+nPx6*+c;CW_91qZf%=TBT+FLF#)b8m=3smk0OS`ANo~w~79I5UXPOgDC
z`5hJ%UiS&<v^p_ZO4<8bl~o`~cU-|jM*+}q%C(*Jx3@1)3^x%1KU#9Z)`5&n{&@t9
zk!+(gR8n=|%^#G_O_Tlg*e)`xH+tXT&Uw}oo;}*<rG)wVm&0V3k+f>JTM~hNY*4~u
z5Xvsv3O9jmGt!_Ei3=B4Lb9T6yr;wpaCO8Czs+&RPIm$S$B8`wKU-8Lqc!&o&0D2m
znCt(vl>P@9Ca;85zoV2%_cfQ=R_OintPI4j#}4FrbABJV6D=Al)Xaa0Y1q!Iz7nL4
zJ^P7WhNz4grUe~kGv#KAhZ6z(0wB^Ju)G_O`oo~nzx=s5S+m^&{G1%E_6bQ90=if(
zlEH{G5*xrrFIe=`*j*RY=2HZogBUcI(#me+IYybW>~oZ7m_juany!!F#d`8b7iFRV
z{(K>kJZM$eUzBSB7{LIK@nzAw8lwPD6`FWV_#m}$J*u}kYRKc#VI>868=b(uvK5?3
z$mx=`yv&IFw-gS?cT6wFg(kzNW&Nkfy9Cv^A8jO)z}!?5hq}qXJ4CXT#5qk(g@xgy
zC=oXN4h&eUBk=0QS8A}<)S}_SXf+e6H-;VTcuBYxu0Ou(V{`4FkYY?|vC9g3D}(|A
zW{SQ7Fn@im&^JilFxP{eHOGw`&lj;QPl^4IY6+r9V0xo+%u7_ad-Mj@pea)4MVUlC
zfhQRfIg-(w!OW=$Lk8$}iTwLvt(Q=74)!eC=2RBr5jffui!|-O4s~y64*Fi-B;@~t
z8=GDxjPSdoYYAO0jO7VxlB$Mp<KSvb&kp+epAsf_2`1K*w#3*CQpUSEUa$7|l0z&-
zVuHAdK?_dz|E|mr2~@u*VNM@ut6N4MdjtQsSU(Y*DG%QOp}aOmo6w?s?qI=r@wG+j
zG(lu-1}fKstS=k>N3PlLnD2zYNt1t~{{VsifW+T>>vt7xzZIIHoNRdMW3NEY{t9H^
z2<3a6X>f<4+!I2cE-;?y8(iBu;Q^-y*<k%%y{d^)2wF8_S+J^VBy>vd`umEdVf-Up
z?b#MeaMoWWniEQZ`P1ZAs;yCDAVHN7d>*F)<9%{CQ9349{YSwkh1dak<`+bK4qbS>
z5!%}eA7+1;_2dekYRjJvJ41`r1<g~BxP~P1#+29xeQU0F4mYg@=?$iZL6!<X3U4zZ
z4>+G7=MiqJF@eF)eDnT<o|s=l^3wz7^DY;&l?+Pz;o^$e=3{wjUTDyk;cz%0Cr%@N
z{0#(kPZQ0c@Sk@`ejU*?dfXi8guOeExakVzkY9~0t}kYiO`J2l{up}P8dQ_62rm}d
zeEBFlYG5vB?LdNV=^z1f@+k*W>@}}iav-PlbD9(@@!+>BAmSd>2i&?;3kc5LVz%dp
z7cCIHBG<$bS@&|Dt@2HWV#o{Ypf4bMpXv2<WT}(!^-w3s1QoR4uqivIs{`62yY7XN
z!=A_<uC{*#c-Y`1#2hL?;&YO=qgzRHk+r>Qr2S=_iZJetCbHpZv%m1fX~JA4T``4%
z%B8Ez0<NE-EnsI`5VM`O9-IbpCC=_ea!%AJ>X-{Vnck>cT!!IysZt-vp`?fj6RtYa
zNUxA%Uj|rq961bC`obnaX&GqfHk^ZBWBQQ7g*#byQlo~oeKZU^BEwfQ@6VFc)ov1I
z-tQqy+6kVk0wX4EiSn6eBdY#eboot6oAPSLJ1|%9l@jXy^U3qUZ2<qy)r6=lTs`f5
z%t_2WX0#GFF(ap;D~3jZGNM7K&|#K}6dN;vr{c0u%l7+7|1}RVv%RTX4^{v#<v~H7
zd2^4HuI{?ry~XdHUb)qF^y_jqsKWP<=)uau-Y6^joO*2_6Awm|K%Hl3H7E&tLWf-k
zC3Kf6Uv9~r*r&RRlKuX=fB`^J*AhAytQnp#9d}&7E2a<iL0{M4-D=7M!;&^@xAb8a
z+d{Vh8l6B39#@6pnGH=MfELIF)k;cz;x&@VS^Rc%78;-V9%lzKut?^?MBtbMZ2h-P
z12l%ENh%DQ=h~wZzxopYJZ#yS61&m9b-46$go|e@dE~El!ys)%#||ZQZ&~^Rds4hY
zlW?67x{N@0v19O@xS0|^-;=2VBM`OX0+<js($vRmb?>gj9fGCnk$OKw`;C?J^2BB?
zY<Tk|1Ij7j6*yyI`XYu7PqFb1!CK*heT{P9pTa_e#c)z*%XdGiQnA&zSNd|Ge^bCW
z%eY+1BPEb5ZzQ8{*MupgkXrkmCZO^5sD#5`j4WaOjP5)>-c^Xqb7v>+#{MCzjFCB|
zzAVhO%%(e7Yo~fzkl)1sJUiBmzyuQs6Dfo}hTz{Hs($ZgDYPd{D?Cg5e6g_`DrZ~1
z56vvV)c=u+_i4H8XPep6=WYL4aEFb&i)WbWFA}=Xy9sqtH=267O5J~=us_>J)6R28
z#iA?g$)XtE26C+1;cA%?HfRfO*$FZQ=0|3|KdfF{(Kbo>JAAKTt5{6U@EILiQqCrb
z@Vjke-p_Y`7<K*R3b^oF)*Pf-;Y^)b0ahU2iUgJlKW)sqiOgumtv3v8etbZ}QkN`Z
z6tQ_^UHI#pT-_FuVDP92@Uz5AyJTOiTw8kS-~*LIwsu<}g}d=$*KEiNXXU{HzVqW}
zSGj}h7mcJ>et52mH_#6#6fAbHzTHqk%smqRy!)ATdW6E`;9$SW?W^p0me3@^UPY|Y
z*Fd26ojf`|+CA9TZ5WY>G@OFi8aBdUx#m;w#`)G0YN2)&Rs`#lg!=tQJ$)nGCbxk^
zQ{QgRbmRnsQG2Cp*3)l^hgXmm?1kO2A`-Ey8F&Niau789#49@LqgNf@CY3CU$cs^K
zR?BYtbJ(>~3KGQX@fgE~fQ>a%dfqi<(&_I3H(?`NajTjCTPZnB6Ds|uVBxmj;$qJ`
z3zzFj*kW7hG$Gb|4C_HwV3%11ji<-HBgs#H;Jt^)gc<&ZD253wY$<#R$fcWmXB!fW
z_Oy=W0_sA2M#9;j=<mI<Kyn|&@Iq}T0pma7)Nw&nf>@TYpJji~x9m>^X!(Ubqx&M*
zp0I)1uwI$+Gx!``Sy9M1DHnDWu^bG|C`s6edca*K^4_HpA62J~zLK9wk6r0Xh`yp=
z%54J9)9IqXgH#xdF*@|{+i=#+HtUm>!#~iB(3z@8^3c=ZEeRgXjvNXvrY3U7QYr6F
zp@rZZ<{S5G&Ep^tQ16#4H0C_D;MNp{$3icXfez;Yg^L-V$Pw<cu$_AU+Q|mT`@XB=
zf&w(*%OR9l{{&p0MwFes$<V!K7Y&A@nys{@T{wmhQ*DOW7)))jmlu^wBF{LsBS%cv
ztmnNyoB1h}<$SW%^pI=|8p|sr_ZR@CJ{>rWD(RpLnNOrmG*x5MdvVVJLe6IxJboB-
zbwX*IslMUUa2WRHtokc7fK_wIRvh#;S`&`94!ID(DT0#?=)hFMa<z~nDJ`{{`EsvP
zKP<BzeF>E`ro{fSw`ioG?326&&5_2cPZw;|MG&T@LcJ&)pq=CTRA5gJz8UWx@;zyo
ze3Y$3!XYTQ8x&SAMCFCRBa0#*z%eEap=_}?-W&;nGla}r%7lY>!iL;&2sCAQA|-U4
z`gnhj;E=_2I=R`oIT*6e)pQ)9y*E$T@$EU;t_u0yhznqW8tgqHGb%}EKSAKfyH?z%
z_<wCmT@o>u2R(B4Z|hNGCpTs@o>nN?UdVoy|Erm<009A{3O;PrlW0Iw|8{n}L*_~G
z)|I2PS*Lf@p?7Z|tO*gOxmF-6UmiFp)WfwtQE`739wgm2Dj^$Hc(y0Rx1!N;o&Pug
zRbTAejw~K9C*h+Gn_fXHO{rNWLxu)z<CZk$A~$TmCwRl}En!wfyPg6aveLLecyvTK
zig91S*M#n!#1kkvSI~t7d627+R&xC&8sGMqZOXY<J180M^(^j*VGRrjUu5P*@PGj=
zO!?$9=(nM{HChy>y4P*evQ{btVV4jc60rR{Z})e8|0aaI;Qc_8&qWTu#=u{&Z}(~$
z@#WBI_lWgC(A-xm#oHX}gZ5eVGdFC#GF}tP_Q1cKT>j#vmkBI{QkrHb7M~py+K}N>
zn)ayOtlkLx@r{vN{Jol1(kw(GK=U^98VYac^adE9@FKs`E;3)c*h+<cM4#SPTC@0#
z^)5`~nxr+j%f*rc#V8{1w2<&@EBCB-5#xLWiQwa}UrD@=$&(%;cOF3;RU|?#3IKI0
z&5{CYu~_oHH^}LkV~VNW_p`*=bBmEO{mWi5m}hNM-19be%#(c9^xcx!J;LD#|Cito
zlt)(pi7c1Q4*u2{-2e?SEcF}_0>3F5?;56Mbu(J{lg3k*zdOnqm%%&VsNC{<w-czr
zg1Gh9Y*tXg77IW0Y|@GhYb5Mjzf8?~Kz#!$nEPIeJG_ptks9H{i?LyuTQ)AX(qh>O
zS#`4MY?0kkCC!mkr`g~2rxi)wy$mODV*H|P*<_v}*$9r8{)m2-J8jMQndu~Hy8S?N
zjLvW`3YwA$rLr8UaZ2~!QAGVPRRBq=S=07Y{=(VydQVh0hsj58p_&qBV6fCnjTDQb
zqoEI;fuZYu)<DAS24`5)MrRE_1*b8i81Z1;74YMiGSY^lR#Sc0{X*$uomx4P$yf1;
zt!&tEn#FSE{0GJ29#;rA1K_VtzP;*i*tAa^BXi<9osy}1e53s*4Yv3D2jOGiQ+BV^
zk9_$(=U}PlU7u)#$BNk`Lp9;}*XOmVCfiPLXSy&-@|E2qoNhKHg)VI216v2$r}gDW
z!h=0=e*X7G4m(=5rUbJh1?|1tg7l8pckad!;r|Z{Aa&>H{Vxw(7U@aXl-$FRhvN!m
z*1<zjp%YOkvbNCE-yi1wBsTA%km_~e%j1;TrD+Lz?XpB>X4thNEy7{b^ZCw~(edI*
zf^WE&+m1#|pg|ifI=2uOk2yWtzDF69EV^b38M^VI@T|v6$CvSyWv=~UEJL;g*J`?E
z+b3nm`<+J#cutZcfH>P=loC5y=0RA`JeI~X2g_jS)(BL;R_2i#FAd{lA3u)O>B7Ff
zzYqnQpMu{vizTQiQFw_Fs!<~JHI&@xY*St%YBxUX;3O}!ZtZBjE%fNg8e#DDjGTwY
z`pE7|Ow&(Uwu)XDLOn2ic?3GjIwf{R=;B>L)v(u6*k-LP_WFw8A`|ZZzDeZvlGNUx
z@b_Pte-NqaR03TzWeJs9DOSfa%Mi3a)STYM9g0IUx{BlXf35f!Z+))0d!H(yu&^6}
ze~`SUt!?!sZ$~k{NVAoNtoX<0$DWJ2(ZLli5Hy1!vL)*6D_MpM>@=Dvpi5F{q6ii~
ze>q-FUwPo5h~561D}16k<+2sqR=hns!*z8AYoW9^@Hi|+F=w5@mB%zB#wkn~Prz{G
z%R<^fk|pV!0lN{;7zd-yq;X^>ygGVR3WvjRjkA8iO%x%zX^0PkeMPKNnpR4?Xk;tn
zh;Z0%Fqm@%Y)yOL-EY&v-@avU5dDiC`|kE~bL6id0r;8FCJ{nK>(P}N>hdAVRP-^0
zVZ?GDCDwVm<oxFON(1nx#rnWTO|p9B2O7VVb`2Vwk^3$>qx4mv;x#uI3>3=mI6^;9
z91{Cko!O%6!+dDgxj0%}9k8p6cv8gXTT*>3q=WL<$YwTxV_h0oD~hmT+CQ~lwxR0`
z<R753wxZ-y%^>5hpi|NZreCoG-5s@$zD;f9`GvV@Vqn++-3Dj^6Gi?N%-cbm){B@w
zjraB6$vDh!_fwy6Rgnpda*At>G@@r$=uPS|jXcDVsLU`sw^4ZMbp&pMa+?0wnjWZ<
z00EM&<E#mxTAK_zt(BqW*~4?HfCw@a#c%-v(3rFv_fZKPY!ca*1l-Y+F)^X8BbnI^
zn;eqahday>gqXCJoIg3|t<|6@yIRP?acC--!cCY{$63L|5rsHQmp`GG{X02qVio=c
z1yX4YtDwnT5fATRQY>Bh<q?bp`==U#qqi+@1DAuQk6eh!^n@YFBWW9I#zoRX(KCgR
zL1NE85yL#h-S2MJHz+qc%{n>$fV?#O^y~YEvW`n+VPToL%k0ppSkK10&CnI?a4e9C
zQe%S}e!rgRDhN2GNO~zGd2D)odM<U74jn68Yy!VUv@978v>rYGaUAh%%va?R7Y%kG
z?I2{#agDa9;vmb5<VBDb*?}T7;TR($K#7gUpz$9xosu@t)Ae5p<^9!)9};P_nlG5o
zHPqF`Kj^Irj_(VAv9>ZxJ#3Q#IIjFQB>Wnc6N_$;&Xe3UOu}(*-ah=#hd<<na<Wv+
z%%(&tlerWR;bl6?1|>ol;c&2x*r~?=hwuh?!h?5(N$uYYY^k61CRtsbD4)1Wcx(N~
zGc>xTkVFMOz1z#^x(_iMT7idFDCht~bi#gAb~gLNrKl;Q>BwoyyscA_-@Hh?tJ1@&
z-!sTf?fxbq_$`v<J6O<-aD}NgC|xWOd>dbG7N;20t>Nz&Z#Xb)Qx2#6c9TRAtMy0I
zq)kGRP>Paf$#;aouhAsIap_27|Nha)cuGKaYW;i@Io%gyK;#w$EDXX3Ffv`}9<kwk
zhr}?TmZ`0n{MC>3wUDe2i9Rr6>W8+F)7iw7VEP#@lL&=ZT9+Xkrnx`%@My1p*3%?f
z^J=FKv9pL-Tu1OPZkuXQtQrJsxA1h9VMKsM(q&03B&0{c&4Mp?1uMxR^!@vnuQc!g
zM=xv2bA~RyWH1~xG0S5iE9~aGEDzHZL?&wB0R&kfO(t91zNy)OaqfA~fbCYW;4Iik
z^)xfAOR+dMjG2SGK*3WCD4cn;>rHgT4k`LS(pXi2%R*00a3LM50rn02uQ|U^OUUUx
zMOb@M(C)<3`Khqr&6Z!@jFWCm|8_w!z^bd)4roohj)r+LP(ZTX>qK!pUsGR@z<|Kj
z|1`Od7+E_KnXTW6OtjoGx?(1>>J#7pEmvL!1!>guir0mcmj>xwr^S<2r4_(WH94-9
z{Sy`FX=FgdP@X*!adIm$^P)tMs1=(D&0|6mEy_eBgg74-V?%CCNKwe;8c~eLhv!9R
zUN9ec86Ws_dlZdnbGtMw&Ql^S&I&Mcep?MkiDdUS7UR6Z(usPuG2<$+>gDc%?k4!>
zujy3a=Zj4QR+v34*(CD<C8W05zA`QsUyF2Eww>FK$q69WU2m?U!*-{9*`H<WI9pH(
zf9Fdk)5y}{eDtn<vm5e;WvI~D`xo8^<fdi_gc_^wnLO3HWFpR|<}3!Ew&HsOj(kIO
z;+fx$+TQ%l2kE(Y^X}E^t<Z4dP9?nsYUS^2T{fpCA>6E5E*!uOc_X0~)zjzUwicv}
z0oYP6#|di<SZC=>E?n0l&ivc)O+^QtL93V-!0Z3@_SIiazJJ`;#s&kVTZsusOSi<B
zpdUg&DW#DRl#-U(6c8`~3kfMjT0lersS#4r-6<g{onz0<_xTf^bDm#z&dz;xysy{0
zSqO6>?|&lnV0tJN=;p94O^aF74CRUdKIef067Sl;4vPytRjo*ey&*6`{hkX8Rv^h)
zS2gl!&}!ek79)OQz~sD(we2#P<N3X%zVUiu-oxZraQ@vras*(H*a=M@kz`MVQcI@y
zf%GcEk1g}UM(*Cz41^Y}#m0N-0GH2%*N&b)ON8BozQA5yNPUS`ra`+(>~-V(xru|c
zBhwd+$Wn93;QLH;-q%1|KCF;!wx#L)jmk`R2nEs#!2eCK)mohr`-W_)Ne3%wAsZ>r
zLfd+EG<=0ogiOzowevqdSB2th?|V?Vq|SzZ2<)tlbGX?0-M!V<l_Z2DAm>~Bf(scS
z+c6JlsWfg2gdK|Fd@mB4PCWR5$SK>Jzklb<GQyT^qWbMl!n_C&uRBXJY9|gDO5lb%
zqR4?yXyARdxI8+gM8GJP0gAOGF*dapTH<Q{83eoL1)0WEp)SSa2g1MJJ)bUt1_VP-
zo7LokzbuRFN}wF_OfOmMEJ78S7q`_X4kjP6Y^vK)1+MVU8g^jig)nBaUNGVz8^qVB
zhY&mF7D@XL_J5H+p7ooC25-Wy_b4AJ*pSm-uPU9zy}tU%%&Vx>TYdgPLILCv-<m1`
z!r}aWuv-a!24Pkil^<W80o3mjYX1?lYj2ZoWxji|3@Q6Y>g*yyPo_dI6+3%J=vHq&
zrM}Z-Rl-u9svOUV|Gj8;Ya5>X<1LxSkVCAy)K8lla#zH3)ZlGxVzW&p#IhW#hgQe^
zJRaGYBIO_JyfLN9yAA<*D2&nj;pb0AWB*b5w=Qfd8vxqV?5mH&pNAsXW<qy{XO1OP
zx;243>6DH-SAFt;w6)z(DWMHb>@x(*!{p>z{<DAgPy`YU1V|92VTyMk843MU{A(?z
z(wyPc0+%8=4aDfjXzC@3{k5a*<c^M7Zp*4G1*L7zGme-Mwt@eiTxE}Lc`<k=ZxNIp
z{<5I>(lNm^(n#Agv|G=4r93#S05MJWFyqT6sm!0e>fzil7u}lcaz)mVduJl2;kdQE
zec!C-kY2adj(KL(bFsj(dkxk>Cq#AKORqi(Ex!uR6pB7i#1Y(^7%zFr&K|#sjpauG
zF&YWQ@HbKAVzq#O)FbAiK~2m>P?)#M05(4AZ9Y#VPZQ@jD%}}-1jP*ks%0^b{r0_0
z*lJ^X{A?mGZA(%s^CP!upUN>4zbmPsDP6|3zGb9p_;d*4DMegAGQEt#{Af+(p#TGZ
zlHiU3!7lb0lOib0<{u`!`&yJ=Oh5m(<{f30Lz+ihzG{``>NNLE_9!vk)XX;Lr@G;f
zeEybgHzul`lSPA?(yxXRkS02~<Pfj95uf|}x+H6OH|3lb`uj$l7|6?Xg)RBQRQ&bC
zU-Ff=ybG*j^Cr_%VpyenYojjUCf0vU1`|rNgxiL{;HNvky(P`4`GCL(M;}m0KB6O7
zW@9CkX8sby4(__+h2Mu2QutLgX8K~QAr0czOb0!U;NJ<E50j75hG!dHFn3fTKU>LW
z&;%v^nb7rAhUAgjiJE|We=4}pVlVG&+GRJ?GVw7ez(Y)I6nmQAGC}gLYW-USDL?vP
zHA?uj4U4_8%jk9XPzZl(CuUK2I%fw;MQLBa2jb8`6mLg9yVjT~LqQ&WV5<6XC%}P@
zV3cUJhim*9Fq5Vmt?SxM>)>=A7I5@Xu<`eP0b^q(4W=5>6td9IdEZNpeuP9Ts$*^}
zyc<;4fZb}qnEZW>Kvz&q-=X<0(nTqi&$ks`^v9GCA7}fh{H8)GbK?y`Dwx;Xs2(hv
z#7`~eew|KBG9DdzONU5dKpe}Y#@p)_+cme^tnQT@+O$Gp<N%VuA^Sp}AsQZ+B(lo%
z_g0QIrmQg*=j3)9cvavF>fAOi&#~a+Zs5LFzHUF&aYvwl5O5l{%?wGfJzt#tpLLho
zMX;7_ip27fd86<xLZF}TF4srz<inTaRP?|Q%pnN7$^w*xC<pyF!!ZFzFCSrGO%W6P
z!z8IsgmG?w`x1k7>l}RyrR8M51l&oODkvL$rjJI|o;p7WMJ71Y58II(<7!Cplt`UD
znCewn77PxkY%_r(nbutvysGDFu}6Pu^n_)09DsrmT#zGQWd^;czj-6MJbL!jeVK1x
zS|gRpRK;JwWPbJe>XK7N4ljno_tb?z=Yrh_pjp86N7hp{A2ML4p~hb)OW;N~kDF(M
zs0`vFIri5WZg!I4tGQ}}Q6UTbxkj}dv^mrJa^8C3y5J)-THve9ml6FEgUtdZRg)K@
zAO^N<OMzD<b3Jg2k|5wbZ^FAzJG~e(4y_G~sm{7KROYoMJ^E4i$bXbIwnbFg#h=Cq
zSf4wFJsxSWnoAKkA0}(GLcBea=T_wvr#@M$(s%pkx$3}01p*B>Xb|b1*ulhsQ@Uh$
zyiw3K7)pYo=LXmMjSGAddd#2?#!Q@e?HA5DnX;L`({!`lor@VH7#2ak)ILr6aQji1
zz684IVu!wqrTuyTNiX{9c$_Gu!_RwbBr~b(FEa(C#U)U635l6_Msk})5CYpWHhC;0
z+V_zVUy!2!1uIpSp#=7I8Ha)3>KylDd&bA&p;L<k3ES*2dR4BO&_oohn7i~i*GN6e
zcdc<l;5JLYE&?W$a`R%kl&Ni1FMG5Mitv0ejY$yUbHDb&<vfaRYvyCeVYuy7@{etM
za#Ec@#w0xo^S2zktX28dL!3`id{|cQ%OeiE_BFW~b~wfbn?#h*1gnc2Ofl3*aDJKQ
zF?bsznwjc$FOfSvcu&2Rw5>;lw`Owx01#U@81ZrbYEY&ochS&J;*}dGeot65s?iX|
zd*}m`9>Z^X)4Zk{subxgSY!l)SkFJJU;8)e1DKV6WM3$^U2vUpo-Jc4pjbve2FzWv
zkV2$K`xaggj`86?84=h6P@KQ5hd(rQ9~W4K!RbLawwecOnnM_j%clzthbNopY_>nB
z@`IG?3wL`zxVX2ff~ohM^RV+yy~nhF%)BZP7eRVNU^N>2k}(BM^KV9z3@Vx)41D_W
z&LcZg%C*p=x)~1wY8<b@3`5a&CpXxG%p61zpwzGk0;j=uEz+XN396%<23eBbsS|&w
z36gHF;izUk+s+}qRC4)$^HKM7!A~DQZRq8v*!`&i+U^;%*;2xylBAt1SzQJM*}(R<
zxzU4-+2o<mga4TK>|{ixR6fVv`l{1c{odBm(1Q!=#`Gc}=KiR9uxg<v2x?xL8PK7j
ztCKx9Sw`@E?0Lc%7GE2Z@-Qw*TE32&a*%R&@~od1K(n6&9schar4@d~bq2RVzFUD+
za(sbc{UzPqwc3?z`aMA}8=XkybMEJUoZFp+DhUpf#&u=oFRTalE%WpUYskK#!jJ9z
zT!r%NS{hu|0jp`Pf}rt9&)tL;#gu(V#7acqE*i|s@Tv%A?Zvh_d7?>ZPd&6DH_Y=g
z#Tse(bOTGbL>DU1@R6<(gymPyKL!A~yM8u2sL1H5r~KO$CCeW(a>9NoY2AOn*y$nu
zy+CV*vWSGIK_aH&x{Y=CPL2M-x^1$dhX}M1IvcQVrnuQFFe4P+T)$JDt34NX<e3Q&
zdxGD@<cap@=7=dVp;zwb@QA$qdl``*8EA45Ao;@JT-#tvg5rZbYI^61%(Ht|40gj6
zt>H6DTkgMfA7F0Lp_lTOwY-#Z)OMs6<u(yRa@~z|fWcPR(&L}p9DM9<YgD^{HgO2u
zxp8vm#Zua^8I+A6bLNs)mmsLy%Jr$2xn}TXw_^+TH%wWM8ORzIQ6kc9jXN%>td7`Y
zxHyzLbj$f2Q5e#<BE3=&6$cC-Dx8D;J5O3DLp$1>s`n!U;!SHOex8Ga^EpFoa6`4z
zVmi7Wxioj3K<Oyy7j?7xP{M1bClS4{sffL)8E1i*8-8*txh0BW>PQKRV#*u^D!8GL
zRrUAgb0hI`84H!kDL%9{;}ySF3lq#if3b+07K9`&{~+%7R2IsOp)0e_Nk02`u1lRA
zTWvs($M*Q~mNQnp$wi`zdw6Tp*I1Mg>19+oA=++7CLitAOE)4ff<ZdtD>l}K#|hEp
zdjXp4Eb;gRs#i_Dzp^r{xgd%$e39EVU-BymU+udE?@lt`V(f96WdJ5&NHef{c|5k)
zRhG0<fZK_!+gM?UoW%u8^1G|wC)!I`xt?DhfbRFp1PgGRc45J}72X}fURpnfi>!D6
zGmPcE0W?VusTd}<6vXeTE6o~WLu~<swxXctD)A@C6hdvvqhY^M_+m9Ju~zhJWI9`c
zhicaxI}QYurYXKn6<xvu{5-ojt$~?ew$*>F-U09J`t1}Iy*OY?hn7`liWD%-sZ(zR
z;iNDzUrlEGy-GqE{{@KT#)l<PaF!A3m&EbKY^(w6?YL;T<N@bb@!GhBDP|Dgt7@S1
z_t`eH9Vr`PFj<?Z;Eon}W6|#Id*|+PQ+CP=Phu;IP-X>D+alZzDSmO)d#C*&MV%z3
zBnsQ$5z2aX_E-kEBo{p>mQ)RM9oiyD-{V6PoQId`h_>2hO`3NaUa;cJM|D)%*Dnek
zcdfy5-0leE9B{%2^aKYCAk-)P4ph2d@nK9e@oDDC(Fo<-C%HQ!0?M!JmIg3TOWf8Q
zyI{1NYk&<EPkeO#*nqZ8GSw$Tz}Xk7j7?bU8Elx&{j+nfC-cSFbBlZlAF};5$ON@o
zY}+Cd=E0j<ACcC!z<BjvLwjU&mT>z5w2YtWudlsLmpA835Upo<D*%Y?NdX2Lh3`2T
zxLWNBLKZuj<R!gvE1~ARD#8ySw-Tq0oA_ITp9*PE;C4)Bjv*YA*>#uvZH0%2-P5br
zC^0^4=t06twe?$<#bMlvhm3zLh@%@452^p;bczo9z}~l;4(>qx{*olN`6<TO-_ISm
z{@(}OeOlpwvi4n$##OlF?K{_bwyqrqxYvy){R)tr{AIwDDobMfy8Q&+uE`23XELiA
zAExX}pR_<>DiU;2x8c(T6Ioh)7pNK+9wafu68$Z%j`Ab>f?j-gn~MlZGLwqDKFfQM
zVrOQQWYPAmd6)^n<6sc%f^#{G!{3c9@x5r*opcn2b(7c1RAg<Z+KuHt)8@P{6B~&A
zL9RDCE^y-tA2W(dk{hZ6>)US_*6|#FR(c7Amp*7rqKOgV*y2&T7x9-qpTWSzlp?Op
z?Vn{7g|c0)JRAX-@#lW^uT5im7t*9E66GB<U;$ZV)|T(Y4Fy%R|Ac(vF?!m`Dw<Zu
zRJ`r40x`J(CDeq5$Q1giP&GBH+Mi##EV&1obHZH=-_gR0sgmqQvN|J0e97${V))9}
zZ~-CV{~?#Y{8`Xu^_U^co@diGR|zViBwP~+Yv#`}`ok0ke{rb{jBg|rE*-v~>lkUC
z+eC}UG&d6+UF&6Y%n<Nm*8bjfnSPfn&?xr0wcZea2Z>So@S59X?$vK1+0M7(8nuv^
zP+Q0d=?^PDEWkS`_{cbM`v@}6Q#WQkE&TECA+!_L3@F0S1itcEQ82@e$%7vhdKYc{
zEqZhMtt)rXtKXSgBJ6Z1yGtpRo!OQ@5jp*EO7UT8VOOonx<sDS=|<@8W<ZW>kOMR#
z1q2+pmW!if-|Zz|>Oy{$YhLHqmRD13Etn=8TDQC^)<5L&$+dBjb6~|Er})hFyQehd
zk0o!DMr;PiPYeC_Eu|S_o&;Wn-e-X^cFb$|fvJMqEr>mP2^S+yPoNfqk#}PPHe4<n
zO}??i!pg6Zyvh~1#n=7KLSd3jP)!Y}=F>I!tk|n~u2^tD9_X7S!1)$Mnc^7Wi1;Eo
z3YQ%Jn!R5dFTC0+356gk*(8<Js!UcLWq7X^GiCh`3cZu6{#0MQaw;{w#Y`}|6eQEq
zNR1U2*k6c^{4jm7%Kq{%Ty{yKJAFC#Fwtb7N<AoqAx)BInTQ7kR8nEtPq+|`9ONw$
zpyozR)jR1AfuEat^BY8o`k&*p0kfH#&poD2ZzGCFYU?kDaemO6?UTWaMcQ{~1lqA<
zqmaLG+PQ_=JKMW_u2uV)XjAs2omTzG>+r#&cT^Y404I-{u_>{mYj@_@s;^4n?v;_V
zn?wigAA3a`+2ZNbP1FdfKNm!opy4Z_O;q^e*rFlxyUYRSi#cV78HoePrZL>;;Boa}
zYEZ;-%0B<hHqQs4(gkTx)t=qZPrT68-b!%a$eo;p+rg$^z>WC#U`>?x;}=q7k3I@s
zq1Y>WUsaDhBjCLWos<q=#N$7aO_$DVvqdahEo+Aqb|?M*)jhn>1*r%i&vsjx;p9vR
zLMysG>eZ@y_C7%0<MNLkgE)?+*~bKG1aPHTQ10eX2BTLQ{BIV|R<KcRZE$2QMqef}
z{mDA1pm8LFfw8&-SkqUMy%naEoL>xafAkO^LH9wtbRlMXr>3U>qMUxV2C&X<<dQ|u
zRqF8&vd*SjW5z`%+lKsZTXI!2NH;E_BTv2-AOH&(e2|#9Vy?`OH{ytpl*5F()#{lA
z4j`!^f!D7Ltjc-n54$y^z0w`Hr!F;El0nC|TP&SNXUU)MM7DZaqS%O}yP@+u@;SsN
ze!h6WPc-RTp#FnPAJ<5?=Okogs!mlmVpFBNLl@{^2@}ASMlE_5O+du;-L4md-c%Gp
z+je;IgGAET*^?^xbge-?+=~NkEoR_x->eESn}@*IDNJh7Aitrs>kbR%j%Dg$S2m~`
zYxxSBphi;!B{pMo(O{ZCW0?hE>#HXU#`kdYFxyVbJ|feAj-a3KLoM)&3NV8Y{KnN#
z-jtWAk2~%1MuS>5Z`TmJ1VF*@4kF!`Dgrqjf%_zh0>iZbD@L<9DA{X&G7wmouxumO
zCUSQbCk~tTg$45cBCcaF5T}5K%7Xvf<f!GnzpyBGiaqE3+FNNtS9pG`^+clqD9*@(
z%qAlo7=RbY)Bljux7Q{A!n@NhKnq16GYNwj|I9_W4)qnk_U~4juxlv%2*p1!1e-9#
z4tqW`WI@4;v3gf73x@x)-k8*og#3|(2<)&WX0&hqe^BQr<1>K+!<gRlauc2qR)s>Q
z;&M>z6b#Ka>&auI@H`AXUXHili{46C<^Ifm_xM=iF=&qlJF`8Ud_00ck~A%wkq52X
zyZ@SN<p+Fi4dr18wM_I<908ts7Hd(gH5)4J>WU^gI#A9x87nsQeNMEq$6_k2yg3ZJ
zLX_DLtGk;%9iFiaNwPZl^(kc*Zz<`mRapUH*r~Kqo7mmtgPK05lkop_GLY~IJTyC`
zcw*0to_j`Lvxeqkg54%pE9R@R)(R|{qQuW<h(N0~fN~_xjgl^Yo3L=ZD-H{w^?|9t
z3Ag`s$k?s=?30lb{x>h*PfoDLVNzKrEXsmTs0cKs8BG>Mwbq0=YWAE^jN#Ab?#uUa
zD2yp<gj;d&QwOOY-kl2Z?N68K2`6EFk$Fl`0VbAHVTo?1$_cYx+032=^=!DFFo-$D
zpPbvKlLsc=Xs{w+3fhY=+Hw2~A0!0^0&W;)o5vNN2id<3-YD3WMiLGV+GR=DC7Qp2
z^uU8Tj#Tdh!yi&yczo8uyEjX)Wl@5K)0~Y^9m9`6hr6L?ACv$0k;YD8LP>s#5rpIm
zD}!xH%oPRp1s+$5v$o~K9XFXQPwxhZJ_QIt?v-%u>{KXhYz(pfG>k@LfD^hj^s<gG
z=bUUiWl)%wlCmaT0X&?wSl19F;EIKFo{_;6DWMsU=_uX-I|XEgAgV6X&`S<I!NvGt
z#Ry@$WrM`OP8Xp?Of%DKP!?bHzlFr?!&cdEm1%jXdva=g9Lug*QHml+YjhQpY<V@$
z>#`+3J)6>QL-1c|YW(OM0q5TDF86lff!aS!bi?xB|1b(Z+^g-@p?b!9xiO>L|G9_=
z!j@PnT2B*Wkr($1=`jAqT2|JQ?FG&rnePAOl}R``;gpOP68$Kz&8m?l%P+B>%kqT8
zC|5!G6{RFx_?vI*b9x90`WDErkvSIKQOoD%aWoV~3u@iZHF-aZrnsj_q!`A<u|V*J
z+wRT+=Ds~o>uqGQGIRSVT2H}HNNQ1So!=8YhG-#XGb+Q4R^YwvQDfnW$5gZhPXW$R
zmF5Ki)W+JH8pa&jf8SKx)u8^8o~{T{#sD8W{0jN&4wCazf3^>b>7AZqE)+xLmxHHe
zSJhh}0{u4DaIC(#8TJ}M@b!-&M|ml~=hAA75%x`a3h-4B?WJ<QyLDI@CMR_Qk?|&)
z24C#y)HBNE(*+*^>6b8Ob5)JUXQk*x=%bEjf6<Yil5><djnPuA2^i$H)=vf@WFSG;
z84;(V*MaehNBtnZ7<gLc`;ln2q;)Zh;W44VL=0}|yoR_0hk^X`&E>{RvEF1-?)tj;
zIDx5?V<AyiB|v|Bt1gHY4y4~`0ry77?G57kllLoqyZIJf(arwYD*sDT2fvPzp^ci>
zjA8;TN`Fv@&x0@RC;htf*cR;UJ9zD`_}qrmdon&mfIXgnt^D8~&kD|)lAsaw4M#^!
zAe|E5pThET=<!Au-GdHHB{<=dmhw?SQ>T(1Wj^ES*28;gf}qQq26EC_z&}PwQ{tTq
z<Sf96id7npGCR+wUjG^k3*3`eQyiPm_f!bU?#@N;`@qX&e3Z`N^EZ(j8)+D%cjhb#
z)2IMT7K{ZSh~LVUotW{(``M6{?D+7?(U=u@B?dY<iA>RaS=3ip&{dp5tLjk~-+&Vr
zn+x{BfR4ux8+=?L^@p^t|AA<o9du-za=5!V*r;B`G@FbYvpP??!sPF}R<-o{WNz_j
zw5TXT@p%1*?y--R;L5wOvs1hhX9ap9;ulIt#)H*5b7gB`%310=yxMjo4Si`!LUUyx
zXE_7@*lN4b^htNZ(zD<46B&owtDl&`J44I!l$T*~+V|ZmI5Hyupoo9)bREt)FaP>0
zHH}t$V~$MIq}_}2gEub7w6Tt}kniy{zS0FTk;}E$0PXCJ0ezD3$U>W@&A4e+<@es$
z<DWN4CL@Q-s~+yL>T8)d%_uK?D<YY373rU%3-ffl@<R#48l}NWW^pv8mkqR?Cd;c!
zR|ysJJAJ;WcKBe}GhKt*CkZ!<qt}*)l}}aw{DQ=oA3CAUXczLxnveQM%5bNL0dG^e
z<!n8NP?+Rm2@o6fDS|YlPFi_zGrL;>ngSz9-mXa!zoZOjdY5VT;HSZiaP4FzRo_yd
zeJ}8A5dbEiPV#>I-LL$;m*6w^ps4fn&YaNoP~QS>1@~Tm%42x<<Vp^gu1JI)bad@F
z;oZt4!5>MPbW4d+^FuhZ;o!K)VEDgsb<zeNaFVt%zRli*{<4vFCU8VYIv8*@J`U|(
zHseNT(COcSjL)H$3^?$n0nUv^-7#dQn{Ts(di9&$Zf{r*yt@K2@KC`5K&MnC8qTm;
zH1-=SF4~ilP{K)@%#%c?V)qIxL9%GyxL7&VI2*q04=Gqh%~BV(L<8@G8(w;ZD|l^b
z4EKC!(Hy>q#5j#<K*zOMOY_2_F!Zj*f76U4>3sp(aL`Pf@caYZvzEz<-TA_*%N}rG
zcNV2_Vi&p$M6`YLo2w1$J~!R)peLu|Z1LqY#?w{D653Vq;)UNhX#cs<s@pvRORtQ8
zapPh`WSdj%;h`ckKAT50>~t%{gLjxm#U5%#m7(DD-xjfs^LEbAixmS7h}5utqxQ_a
zk(Y227?<D7MsmXviWMP0mtZAlzTSJX^uYFGy+ccx@1PW@#tMCPgY@iTEnd)S7A5b|
zZMa^CVx55W*s&DeMlpEzv$#{y9Pkf3=D1T#k$xRf0--<4b{1YI*$V6kuIpLXHVZi(
ziNwUB93UyeL3=;4*gIrBmf|WPycSa<k7GiCC|eE~%8lC4c><nI3Bg#_9)0>`IU|z@
zON$z2^D=@9wBuu1?c*o!6RBoc!7xrz6cQ>}X^yKk9F~0UP#^Ro^M;_U#kB*h*7m&g
z_+>94Sn;WCD1?~j+PkUgHQ#Yi5&Ay^Tl|0afA97|bX}$Xwvf+9&XzI%6)g0eobkU=
z*V=igMi<6^eEyUE?*s{^Ga*#&+-M=SnI$>HYbeeOFr4&&J~qLgUC|KX0{}Y>0%R5+
zq{m|@0KNxmlu;K_fHc5?X}eLi)C<d%FV_2Ot5>Qe^l<N|gQ4#kg!-_T8Ga94O@q{S
zSVN(G_yVofg*mi!NTEVxtupt<vKtrA<N(hD2m=DoHkI`Yq9h-y@5gi<g*N+F2B3Iu
z01$+?%ik}Yruzjdtk#V8gy2pqln=}3W>@z+JnND-KSCH0Qgw8z;<qKtsXJ?BNrPwm
z7KDs6n;Hru+bCKg#ZGZ@!=aWgdFG7m7^FeQmbgmyDr|>hok8Uh$>vu;Ej<GItTP>I
zrm)gw{RaLo#GW)ijhWDHk&%?dg1lbZZQN08o~=SjjtaM{&0t?_QGVpVwZ#JoG5}WQ
zXUecAN867X@q;toOL`E1^2~YHrY!K>hn>@(@2ZW&8reI9W)up559Hq&xM<;NL>CDw
z%x4rKrCjCQ3m-nqFM(gnnb^wcg65u&M=7ZD#MTD4O;5oAegJLJ>F*nqfFi4YkQMQc
z7kp=1&PvHjub8sRBR$NW&{|-7HTlK<7#hCqg>mCAGjj%&qgxD2b*Y6bwK@!V^Q?7;
zO^4^T;vFyPXHRr}4^|3kO;iWDPgA@$r#|3vEp}}lP>}qc!bm#$ArD+=NPRXz@BI@?
z{qJ{%1m)k#4H_+jw%Rv+e5!Xy-I8iH)QngJx&549&dZhXP5;sDoAH1nU_yCwzSwWB
zLkZ%7kM8?_+^=5#-RXBrlzHG%AQ&#M`mH&UdnH@VN>=NuI~@R`8L1Pyw=R?KH+V{w
zU;RuIfx^sX-|7tI*zq3PGssJA8wslTsP);|WSf|={?qt+zS>LIDfQJ|p@&>#JN`tE
zo4Y}c3b+Dwj`tT5TdtTahMvhJPFFu2^=r1C4Hr!F-UNR#WtF^d{55pT7-09b&tIwa
zaPY|DZY!UXGd;XE;s>!NtNzE2ztx&Xcr1JNu6yw@J<7horIcx8MSC_zg7!x+o|Cfv
zz{oXVal_v%@XyaX-B|fwE6o0Fwz!bXoj${)xtfw~|J}y2slBy5qqRQ@5(?5T>zU2Z
z9K#FzQg;8>%8evY`t=AKV>UE-*t9z3i}N?NIC}oG)z0L&#@DJf)8re`Drm3iz}DK`
z!K=?#Ezt2Ml|7`w1NdZFnDd;vhw{cr2Bo#zH4$UPiep*%P2Wt8O2ZT~F15V+C|YbJ
z>^m1y7);DN36zXow8GA|CHUG-{jWMdth-~Qau99Su`ZBG!qQQW%i)F^zR_)Hj&THx
z4;`ndT9$R398Zk!Bv*HpUai%+85^ul>@LUe%wH4uBJ*F>`&^N_-1WoXv6av3)}R|!
zuFVK08sl#!uEDRpRLF9BmCuS}%X=hT)c-JrMtXQRnKbT_9CA<beLMPAN_33srCTFs
zM@NhP*WGw&Es%7GqVY<80%55`i-`mE&^)mc+3q|ZLkahtt!mmi9=-ni<|OE?_V?JC
zNItZZaAXXrc%>bbotDSvRgd9PIAYu;0Pw50^xeK4vrI0u%Ye|_YdZPzjt+FMfGhL=
o%Womn0>q-ypZyB`|NJG<k-t57b`@AToQ*(ROJB1PYy0H?0MaF|*#H0l
literal 44322
zcmcG#bzD@>7ce|`Szrlq0qI(i5&;pAP+~#4LpqgILOP^Y5KyF)7LX;Rl#-BUG3YdC
z1w}v!r9)!hi{Ib#em?L2@3Xtydzm@s%&D0<GiUBYeO<Lvl&q8hfKwXk%7y@t!H*;a
z6bXELH>;!x--!C!*Hz$~_{qmyTs&CZ++5rozGjOD!@p*@2>#z(Jou#e<O4rsn+FG*
z2gBE9^KAGx;STtMVa$I(!RFb)!Qv<N!P&vt@UJ;Idm^~GIk*|V?q-{dXT$ix;@SV;
zPU^Fp;cs>`e4WUiXwII1ue;5`;?2S4&B4Jia&xwMb9V6mK$_vt1Ps_b0bh5Ucbl7c
zCz^M6C$h~avV$kGVFJM?G+-c@3=DGe*@eFoySp&a-R5i<^aSIi-~<mYCK8D^_>W^@
zRZw`UoBII35p(hfCC;xq0)PP;%8J(mvsYSOYlIuUc>RZUK79Gc67~F-Rsx6pq?_*W
zTot#v&iziVccQ`C6Hlh)3da`@i8J4umnracSSRlFlGd>&<GU+@n0viS<;Lo=ORFc^
z57cOhuYm7Blnf1DaZIF0_%^EoLGUfm01ID%&6)pORlI-(^#1&60>RBAfqs6lB`F{<
zqXG8eZ+8X&?BV8ln_e3O5C`*<@_!A`6twLvhwmL99UbKok7riM!w!zqbHz))7q?gE
zv3qq~ZMz#z9BdlDBi}rC6mb0b&dl;7^8ttQko}llg;49N$2)s>`BvY3*r)$8B?v(0
zECqq7J*0{lt5EUd?c%qKVUCKAiW@VLv*dpW&NrVagAbfDLxtI9SN;_IC$SK;r%ldj
zvh!`T3OZ0t%~S)2*KAvdxWTW!FaIAvcF1gxIMg153_o7t+8O(L1A$jO2kk9l$hKDJ
zc6E^0XPdGAsm>FLdEibQVBC)eE$zhgKdsEe>EV0p;Zlts?qFb}`a;)0*iI;ca^|HX
z21PhHnAz`YyG#5ylVSDCbY$F{8U+JM+Mx(if1qz63%Tuw-!booh=+U2+wYw7#-quP
z7jhK>=#Tw!M7KUmdapHa_A4AYiSMp1QimU1%sqYhy~?I`k>M~sH+;Qss4@Jo>)e4J
z7j0Si-}h16=WPzYR}woa`wB9K$MwSY1$Mf^|Hj+8&8E$S`gm_IZqDSqSMH3Y;$0yd
zTd(<Jdis7k$286i*4t^yt3&Lj$=yLV?klPPQ?~#W!5)bTJa_c^7$d4Uu5i_fI+ST-
z#A{)weLQHL*5HbebhMl;IPUr*H|N2YeiTQ{ItvtzLd}?VYgP$~kE?Z#)uFu!7Lvac
zyW#u$2HI}Z?fP##0y%7tZ=4rsTA??E)PrS(Ry5a2v%5Zb+N=+^mLq^&+$#{WfD#-{
zK7(_V{L!Px$&uq$V}7`EWSf3X0qL~4(-NWg(!6aWAq7dOX&RlEtztds6yM3wLg50|
zvW85nnS3t&widV{)t1l@VW9kz8{A4%i?T+^XZo~PjbD)e&+!cyag{eE<uaZ!;)V)0
z7sHABi+bVBj$60ZPuzf^ZAO}Ssp+uT_MH>eP6)KO{--TW+uD}OY$d#YN7VKVZb*f=
zw`*HT#lF*YxEHY8d~iCeT0vy<Sy(Rfrr8?>HR_+6zh*ez>vz7UdcLAM=9O{!OW-Uw
zFi#|pdjC;2S!VY|d<z}^KR4Y$td(~?q<Xkg=*{|9TU)E>PGwQwjx03Jyq}etlTaP2
zrdNOgxyf$t-NILmCy9GYb}i-P%2WhelDo>dotq+^t^XRS{{Fp_gbC^nk`)r-+jAJK
zi9rYQVt4B)0q&9+!ZT>!XP*<hXA<7V#Kr!KgWIVkeKiQNW}UQV87_p!0M4<nW;L}i
zX<gE@>0S5$Ly%CAh#I!(e=jXlN33PD?g~2fuKPtnFsSXi7)2`wt}Ob?uQHHB0C(`V
zjl2r^B>XV^_~Y2Kdw;mUPTFL*x4zPxuqm6(nX`IVC|}93rfIXRgq|i**ylP<{4*nt
zs9Ln-#y+cv>a1j99dFeRo-AW;S_hZA9&YW`iAK!+$}YwhIv4uic>w@(<Ykl)9gcUl
z72PL!Z~YA5TE2R~TcKZ{PK#@O_Wl;$W<hwjPH<``;6U9&2}nlUa$_wona<tYeqn70
zL_B0Yb4XLxV6uXpdJD2j%45|tB>#;mu#@v19;l?nMCQ!xsBbZ{j8R}6E*e*<<+4N4
z;~_dBrLneO;sCq$e&i++Blg>pf8_%L-8|BM_A%g*4HI7B2Bq2XI37qUFbak6JydXF
zijdhI3tz78JrBI}BKX1MR(UyGUHe8QaVz}z4l&e&+U{VnPeG`WMGo|e2!m?2qcZDe
zhm@|$D1)rl$1-ODSffm;M1ED3s6ni;DD0mOKL`_2m{h{9-$m4ey`7g`pzUsPCuoyl
z^l~J>_nLXww1cPjk=Gxk^UWF%YB~^8$MEj`+}Ins=1?1-o7C`318ub8+9<DaL$7lV
zSE>h17h{Vb=#((^=eMSSz?~OPkZ{~@bD|&{;_eS(-%$H0guo{u`~d5%bUNG%w$twr
zNt=P!&6L-Avd7G^3J1qxOfUcQ-0Uyd_Vq|~_-1H>*v?eWxv;-6a#ij83F`ATnefO~
z#k`guli#{>__6Y}?{`12c<*`3$G-)v=FpzofdcWHS!}iixhE|c5}6=rXjlhzG`JJn
z-{b(HmGzsCO_hBAGyahr&au79FR{k%PIk<d!}Vpqpg)(~H_gJDPQ{DJaJC$pZ7hXt
zyx4ufbj@a9s_(br`h&p3I$z#3`LUnX3)|J40R^2g<$g(92Kt5|4SQh0UPl)w-Zqx{
zDl40e6>eDORLa)g^QyTxsZ%NaEak|rt^cQQzPuO_+W5!4es+k8-YVeH{!loo*!{P-
z-e{@7R}=BBDn4GI68x(pX2k~kCoi?%2HVl*LiGwU8!B?PhAD~ZQRDc49!Y3oqm<k!
ziJg-sv}I383<k9}kEOc(=Ye)%^n0fK2JdE&$G5Z8_rB<A+p<q(9qo_(-aE=k_jRWC
zexlK1<FqC7nyn<}`PcL>*hCW7FFO~G*Z+hs5ADi&p$Mj3y0<dna(H0pE4ruZA&b=}
z(Fu|jS&{aEAAZPINN&U?{x^TEA!n;A4`1c~%uO`b;6Ho(!*oWWzHRrxugc?{$2;Sv
zh>yD1lNCh5kHp5mKYQkAR+XL=RvUVHR_vw*icsxH)F?>I`7?BfUZ|M3{ipq_s?tY<
z=cu?DZk}!T;?`}c%Txm1b2PUZ^?T<GZggdC-u{`rxZiz&^S?p2g8^rGFXI&=<A3wB
z9uHXDI{4uH8KU~n(sPZ-*)FyIdHzo+3<UQ7^b^Pptx2^hlH=w6BeJfHs0yQ`KAfTY
zUp*h>sjxrazLy+_EBY^%=`^AM!~dJbZ~$ojN1Ox!1V;W%G4LdeSN?BY6o3NB;TicE
z)&K9Y*fwI;s(lgx6L+dY0|EF>IAtIVXiiEzwK7oP4E!(~R>A<_ArycHem5jI3FSSn
zUZ=!G!Vh5jj4A?<p471|Aqk2nMZYO8K)4eM)6~X@S<Fe%>!?hCJ)!VBiWTzwC&s}W
zU4Z{rrxXRGPa+ON{0#x14?p0pDqaLk$8Zb`)1XNEAAQ<S2<)1|zYtYbl;Iz<^oZ#c
zoC#1rj!A|<2KqP2mhu}DFnbe#qR0d78+QDZnvnXMf)zC&xCO2e$-5s$R0Bc}0NT;U
z0IQ4iMI`+Q-(v>237iO)aufmhNq{I5fv8b}#JnZ0i2>jPA(D`Ao4Df6H$*g`YeK`!
z6Xa$kX(AD59L(QZi|1A1H6Vb6p95_glHcKh>Az<~=}Hm+;Q=i5EKBvE=K3BN!t*4;
z#yQ`oVy52G%U1<qp1&W>$RF>09Q)m;uvIPv0G8}5@!avX0z)uuny7$*iT0EF0%_O)
zyQ^1Ekbj}?6FZ0!Wu&vtGuPni44PtqJ5DZP1iPnB>J`-=H!DD>!(8Q75K<Hd>s404
zY3jFY)`G@ZHy0wKxc6h8AvkWfRcHdRD1mny8YPW4cLMQcNXQPZY5UuMWP8X6n{Oy#
zV&~KOxOK8*9AD|i{SpS;?o7ELEDz4P(=Ds4G3ndIRHf6nsM~M=d;B{*8rF-^+X2bq
zqS`?!Ilu^Wc+?y!04sO#epZrhL6ps#kA&E}&|crrSy*lTp7gIuOUOmlQk`K=U)Cs|
z6lOp88BTX4{=)pLV&{_e6j3yeEBw?+gYQ#`T^~z*n4mqY^NZz_*T`|?B#>F>y%VIA
zh7F>peCW$#?%UgY?bZkkkTpXQ>;(m2bv&c<p~SMf&O5Q!Kcl@23q{zpZ@Ywh2}0i4
zBG_za9-gdqFpnn4pa%vIW#NU^iRskss7m#UfFbOFh5zcI*Q+%?nLa`MP=@{dMj;M{
z!P-w9e|&!_y(2}C0hgDW!+zq^E`=GR?!oJNCc*-^{|b0W?N;fj&zWQ*a76JFyd><6
zwt*5|fFw|#U>FJ*xa)aaN9gLxQk^->C4pJCITOdUSBk`0%#lJR{}oiIC#Y1cc2|6#
zfL)Lb1IwK68AXv37vPQ}48R=XnBKoCb>gFWhqwA)!I<*;g+J`Qt7?^K93$K=!8M-;
zyc`(~qlfh@3}BS8cjp15_96GW#^#H&9Zoumzyi0ye4;Y0K_d{rewEMCn<7BV-Ke0J
zWi6auUj**$1S-PZr>LO!Mk`}4{Ux$EpNtw0R5<BT^*W-#ETM@_)c;ec>Rz^>tsTSq
zOBH&J+Yj43TpEJ(<s(6hP}=*QLE_P`<lw`_kDD%pO)CxFGF0w&i`S~_7Z6J4Nrz1t
zcCVc7bw<O|^lr(IDbTDmEEb$e9%22#c|EgKf45VwX7zZTxg61O;a24zo{h2<qD8~f
zL&bd>#9FMG?Uq!<%q?+g>4u0a_kI7MqJxabH;LaZOvU-I6oeQLhns!(Epy1{uc8Qj
z9sAl`!$-oTgsjislL;&fZ6kFO*Vv$Oj+g6I%|AEA#PaBIIx?QcoA?lNC{WmHBD^@5
z*dNcmxah;xl0cShlfXHMzmlM41P!DA7(cs^f4%Qh)Gkw7tG0d{EGlEgO1mIdfa=I3
z#Dr~6`YU!Qak4YnO7b4tj!yErix9Z^`fndynd@GelreR05E?gz5C*w)m+}_sN<#u)
zr-v;PTaW3*-_i1g>D0Si0N<N`M$k;o$>E(+{$9RQib&bSfXb-*K>t$Sj-S_ax!6Z!
z{xm_#K#ZD}ZgANT$*PT@<PdtP8xriB7w-70lIo<5MK#fhaK7WpyJ6PK%jU5U!rPgh
zB^JWJ2}`6Em^NqL)&Cj^WUpo){5VxxeFaRFcrM45j?Z;lcX3ZWT^yTD7(KkmZgr`A
z@X@pqz5X!_D^tJSdXTFPnH3=?6fETD4X2V@Hyw-9RO;~~T5D>XG`Rl6w9i2g51dJR
z@4@XyTXqGE86Q|3D>&lsdRRA%z1!&i-F{fSmqKb2L%_sV4P6U|D1Tu<L|C>vIX-Xd
z8BGewSuhkgFLylaeU^A}BHsg-A(A8YbE*8WQTvnc;<wA8Eh^aBL^<gnlcpPEz6Yfi
zr=jrfCd25uAZWcFIj@8wh}-iL<NN0i+jmJqD4C*zkLI6!&E{A#-$_urAO6J)7$9#h
z77VC$OqcaiNNSUZg^2pq`g46~aIjC8ml0r60?$zE{)2s!Hp1%%UvQyjW`93L&9%Hy
zBZ^tfRz&M~8nJri`Q|Lowz%HbJ=QXJlWO|lD}1X*hy9=y?u<|%V#RiEuXc5@rM1?y
z?XQ}kXG_b-oqmr>HBZlA-O~BSxUS%3)ZPo#-HEG8M*H4RXYhOvq}nHb4}L4#rRm7U
ztJ|dNXnD*{IFp`8bL6!$mpki01(~VfLVinn2aG|}vo}oT_eDhK@{O(e8u;U-!<ncT
zzrPMm{ULbNwNp?HrZ+p|N#?Ji2&)E9zu5DE+z525l$Z0>pE1wA_+^{!C%xu+<FMZ_
z(E6gIZTr3E*2MGSc#a05rSFq+pdF~UKsK)o<Dy^P3Gyy(1xN?3wuRjLzCl)XiH%}h
zQ(>9)X*7erzVz+uQaeVh&NcYkEK#K^Wim4ynH(uhXF<T*-o;?)>}ym0m+*TOR^(h6
zLoe3$D+anQ;(vJBy!y@DwMXlXn3c~T%+CW`zB@*^4Q}MD?po1iWn`=D*Lp+T^4kj!
z#D@gIdnOGO7IIdpke{5p-gOa0kk+@k52jHegNYBCGlBZEvBCX4(v+4#tx*noN?c{p
zGGloMxWvNxiRtQ>|8cW>ul=Q#>a;55S{#DFf-s}sX?RaZ9rssljbZk|-{z|D20hI;
zm^EpA-4n{UIi(D7v<T4-cDx2DAw=#zYnbrPoz0@#(eaqhd)_dgk|j61P}nBiO_YlX
z6xgnP%yqwkuJP^-`a~~$2nx#J>S(gz^_vnRuefW`?3A36RQ&ke#%m4HQc(CJDwQyY
zZA2qP<=h&k-l>0nJ9#O+Dvmwz{`%@K(PH4BWb_k?`)9f7lF;p$QZ_`=1)czsz#i2b
znoyOdrKIHK(Jtb|Ek0FQzID7=xO8JRL(4*aflS(_$spq<Gc&^Tw7$rFpna-_<xtY3
zjB_-hu`tTP1BibSoz?+WY}3*w)&0~p6o+VB$oU>N^z*DlFAsi@*338%aJ7i+gm{0(
zBVw*8c%ton_Qfv&=V<1nDoQj?@x;3`n3LBPkakMFnK)t$1;JO!-dkTDCP?*y5dSds
zz2WDNut@prf5n~O?F6)1DsqaiwM&yZn5?+`nW0WOeHPTbaXMZsZl)rPixBGedeq4g
zph|HcR>_4Ggp{iPGe6!n3+*KlPkNd+)Yrh<n^^jU9=%h&F=*T^2rB936XEkZjREX-
z=vl^p{Ll@GAmA{D;o3~}uP$ICLw27AhfbkC3yf|ay{tIpnq_9LV=Px;UDMNd_7CFX
z7k%wKV2`w`UH(JX`w)$zsJ?@BdppTzVq#>R=sM;W8zB;+#3aYI+K&NmuxF5V)B^lU
zfbU&3R88S($vmMA)8g>?g{IT=*6QmCbdELvw;0^YWGCV?xK3K7zcIB(43yR0fi72h
zjhEfEy1HQa#yO=wy7TR%8e)zCD+I`lJ5f016t=Tq6UXc&fmyvvUtC^BUT-hLYqBpK
z>gsjnJIf3?+LVCr{rDw%U$GH(M_6R9G}=!lLca0~jk$nXfDrQf#$A06*wP5KHh+_k
zuX%~Pe;{gSNxF{)Lu(>9)oIs9?5kUB5gX@_c8p1S%EIAU!hB}s#&Xq9WeYOnJAHqM
z7ryv$weVWPIKd@%2A#rn79?ZVQDd|X=Z_MZNivKeC!XBH1{K#B-)>sR+;z@=n%-nI
zPA2ItM)3T!3G^%!oe{4w;hcfmf&)?cUmO{1DHqNHAIv_s{}XB8Qj{HCzm~$a>CeG|
z<l<OAd<X@ULDFOTT9C3m6y@>(Gu(odPz6^`BZiib8x{BI5TX=MeW7Pqu&_eXU>=DM
zpn&qoMM28N6TWITuIr6Q|7ah75$JLOX{SoM=jrom7-Q>ajFXWIC($2-6%ipZ(Aot0
z1tBh7ZbS}pP8EUYtR^E~P?VvEX|tkmtK2xfIi#HlDcW;n@xsr1uuBU<MNl}p6sZ%w
zuX){ts(6U+@k{r<TrA+*Od)}|+{7kr{JC=XZ+%HnmC3giS!-uJ0XD)M#J=F&aHhlE
zV!-(YB)-!<Bl$t1&>-yRW7i<fVMsr6<2=%i@<%SxQD`v}X*Y^NM<yH|wVG10jV4)K
zQA$=z>!FCT#-UP-&H{1#7N+D3Ap`}^Vkm=@@}sX<>YnA`OBrB_T+QuqU1(Y)pEp4f
zs0HLHSjAr3AoUxEu+|G%l%VykZ6EaQsvpE-J}AnHC!IGvfl{ELB(2)Zz+ENY9Zs9Q
z*J{&&Zy9)ec2l|4T8p#Um<gBjD=z=C&-7-bw-C~f%=1|!Icc~BaYh@OzW46iTpZOi
zrBk;VkF|vX1*yPRM%wAb_}f}K!(u4kqPNUA&J<^uW84vYLnzOR(xeK6+H9DA;5?zn
zEbF7iCv_d%j$MUflPUA(znVT|9!u-b%lg=(HP0W3a8YbzM2JEuK0gBy=&)PIdkpQl
z9^bLyX+j^6)_$K^ikWcsX1hNnQ)9Nd0>0#2to66>th`}Y1e7!5thR6Ct*9Von(f%V
zd@!7wsm-QcDF6AI=}1%m=0~E)i@qGIX3aRo9P4w8k~QoM4IkAYiNi#B9AnCzv*7JW
zEOFV%{fP=qcL@3M3?}<?*iMhhyyyy2qCt9);o-;SJ+4r|E3l{pRIeR59nv8@&mE8L
zdVSrcWxPP!pvHuiM}%-jAOx^7ObAid-CqC;K9pGS(#Ca82@p{28dClx>g2~`xD64c
zodgC<Xm6f|v|=oncc?8sZ&4D;Z`?F1`>Ik&e$szNT=;+<8ew5Sclmj}8c9ywEz>U?
z{Oly98e}IFcwh>X6u3sj79-9+qqb;zG{YGD-9<QX04K2~61yLHOeFJ~D1ysRrE@Rm
z2~tcTVBcOzq9lu>)z8mrhz6X&lJuscaUCwTsFhNr-7|SC`|2ITODODlhCN%-ld6&7
zZD5AbZ;f=ex;@jMaOlHc|K(6Fy}mT<%aba;&sv|G6gRFfM}ZP-Bit!juf#40;%;5E
z%Le^;&=tQSzxJUBQ&aT0@a{uRDfMUjy7j4#Y$#Ls&w_#xF){~s@}&!4)m~W4>IzT!
zl5ZDj$4y(&56<vP7!RE=?peD6$;JciVN%ppp3advJpS=1ElcME_Fu@^a1H@mB@j)5
zAoPf)gWB>0+IBzR%&>&Y1=&4Y^4@ebZhcr*?mUFi9bYG?L(G-nNYJj1?xkgD>Wh{-
z$q|y_dmFs40<Y)Y=ONNNdkBK`tpW!8B2}G6^OBLVpMX$Sc_9rM&0RVDiu)k${@-63
zD0oXl;@XT%>Sb!@RvCl(XAJW{zOFx-e<~Mw5qV;-8zq4OngIai9>@qW%hH$J`DTn5
zu1xb2=DysQyDi6orS5%<#u*GNT<wAE)N&=y-G7F_@`j&<N&>r9yNSXZKJ9eB!tZ}v
z$qVt&4})8cXa&-L59NXZg*S5%bVOzH!QwPi)J*1R)(tLglbc6<J3S#5Ct(h?lFo|6
z5HWpI?<0v>!jvaGD5t1(xVp>L1h+Xx8h+-YBp~Lmpa|&eaz(YE!6BlU4mTHK%fS3$
z6TN9MXYrc~pS5$|&q&i!wos3}*9VP*N<t$9NIiQtrA|R!A9g?DE+H`Ub@g>_vu-G?
zp8m5);H!4zbDa(++95}ZBfd`(nU-}~QB-$}Cnb!tPdqdd5p>-wk!KMYQUrJ3e!nEB
zyAyB|n!WFTK#8-H(08#&@UkD`zU{h9@mA`g&*9EpO~<OwAOBv*%aTJs<M|M1HiiIb
zFwfXjIpcZKl91xv`1lWHq8^Acw<mKPO+ax<)Z#ZeG>h9{N8-iZY(C@*C{(*>z9V)`
z)0kkRq)ziQB0Mf9RCjvcRVe1EW4AZOtg)HS#Y$9p2l<Umxb3}@KYha5mA5}x{P_1j
zS_v8A?2r>FDppS4i+t3CW;1Ml`YGQh{lfD%PcUcTj?_=+U593}7#vQV$i5ZzC8b}F
z&6H0_Qx(epGtK(hQ#{lk$~RSb&2>eE(8p65b@d8*;ihOBk(3f%Pk;C)7V0C*Cl!;S
z=m!X`DCbvX8n@1W){x=wT*lt_TG%2W$6ku%vNTY7DnJhqv6BO|Q}PXZh!b0c*(IiL
zOom*7Dyi%)(_g$HxBlX~YOBeU-|q9vJF^B>_K>Kk?Kq@sjzHo_{voJKdZeOLNC{qd
z!KcOrgnrxgzCpTMW<PvT1e-@`jNY^Z`4gQ(zm0I-u===qU-IAHh~x2~zn8sr-IT{R
z$~8VvJf8K5hqTPYNGIbPZ(H}b%`KA8eOLNOg>#eEv}3Z2NK=%dK@6V~nxvNqauQ3Y
z1?i$6wZSOsEgx1zZLzncgvQt4bd%3=yY6J?lHdo~_ONWHVf)Z?`W~?7yk&;ELu)Sj
zsOmos`9<RW)bjUPFu~b9#o$5wGr=D?6P;->`YJRGg%$q~90PT?T%Z459r4tmZ*(EJ
zlm-l}v#;z+(wa4qo?4zbtANC_2)tGVO&v|!RA+Fq8KI`N+_Mz?HiKi3{>S!WkcE0l
ztbcK9b9#vm162MaDhiEL+%OMn{=KLF%+bEqQDreiNBcDskRy^)&T1&Wcv?SbMJvXB
z6f$(W*By<svAet)EeBp7=h|`V%JEqR@m3nfp!DOxFy(Mr5kvIU2FwNZKS<Hs1MVfL
zo17fZy5{;k4H)Ol(wR{QzG)NUhuik51-XCN{wbbm{G()})c_Ot>~mEgunszAih}D>
zWW(ZQTP4N4Rj%^;Oou~YUK)kHh}wI4Ag5d;$B~tA21MqUYiEt}`th?0;2i5(HJA3<
zy}6ft+~K{de`f*O5a*_T79TMFb2?Ikj|=C)SjtX$Icsm=J;ffAdg0|)Vgho!FQSQ4
zRD?$6ydWb($;=VGU14kxY*UI;88K5N=rpVJ0YaxT2k<#QExomZzD@x`KfPKoc5*z@
z^ZVqvMG{7$p8(r3;^x|=9F<y)0eFxd-L1Ox<-W>LBIf1&CBbW03aF|qxIw*qpP6zL
zswJ71L}9tcqrQcHea$Xt$ME%iw**xt^)hXB$)?HLhfD18M1Jq7>&LHD8@I3A7DBp^
z_qw8SB2^EfUberp{lj{dA^g)?O_XEvT`Tk2=#6uhrE`p*`QoHZz!sRi=go%1lb6}P
z`En(5GoM13)J~rXKkwnRL$a=|<sULMPh4Ww_YTjOX$H(lfyj*$Xn053{RfxVIUj1Q
z0+Vfeb^Vt40x<6kx2O(}CNgMqT$gcgxON)<X|KY$gGhZc5BOpo&qRl?j~@yN$R3i#
z_K<^Shmi@!<hZ6!{)nW@IbUux;X?9W^`#dNzLsAKg3qjA;fQjW^KX^Yn%*9@db}uI
z=Xo_0b}F7pcCMo8;)kQFpy=GG$2TUnAJK6I@T9KNp^gA?6B~4bwd4|8@cnp7oFMf^
zsH`HT?2UWh50-!*BYc-xc)^`0si4r(H_N6=q|XPJX}7?2e9I}wh33hd%pt^4^yDmQ
zoTda8H2HyuLF`-;H9gQp=;Z0_GQ`Ctx0lGT|Li&>!%xE)36zjJRsh5;rfT0XSe7@!
z{^&Fck70=`dLxPXmh<S8VK<ed9(J4qvte?0Ak=3m!(j?e+|mPubse@I6&$Zw;z10j
z5I+{?nt&|tRhhlno5?&MlwS9JkHx90Ren24wD(u?rHbS#;M%;3+mJ%yFGo`We!l`#
zQhwQL1XfoBsLG2L&nEwCYun1=I&`Z=UF>q7C7r_G^X;SpiAsRgK6ljV;nNJknOPKs
z?`m;RS~%=P4C_MCzTo?%SqBoIDIT)NFDQ#|krD&}<3E(S3Iu*y`RR>DMaJqcR56YV
z?4wL>LmgKc*b0Zw^sRlZPkDF6B8cVeIJbd5AQOmD0v7hZD0L~M+dE>}16?-K?)J^%
zn=C$ExCJ!0s2a|>r#~GWZDSSlAW=b8;U^pv3n1|_WJ8D#zCh@8hry3(L0R_{kq8XF
zkrdzkCw>$voMt-YwLLa0B5KFRAmYP5@spqQ0V8xW6E!CDUhxA^%2e@7vwcBKP_C+n
z@~>@T)_q(w1-9lnp_D2<+o{<hV?+Um{DEzxMhu_n!WdVtQ(?Y`SGSIA5l56EX%5h9
z>5~YZF-5!<lng5M63~$(y)H)jKv(SVy<FCmE(|#J?Xrs&g=k1o1=*<rg}>Pgl;~YA
z>4#$HqoSnpzrD@}wSvaJ+gsz5m%Fbo!P^FGN5gv5y*JP}|ArRxzEM5CsfM5pQQ4ax
zNlR*O3PD20cRoYRuUV{W3?0+MWw3*2!Mdv+$v7mBVj~q|n6l^bD@_hjF7tvS6KP`m
z)1^_G5DAXUJ<6gM*X_G^?H0Zfe=L)_nsxSdi+F7nC|nS5Rs>J+<5r*P@EMk-x=Qoj
z#~%e#?=3z4lZHsD8Msb)<JYw_cCldid2pt*k(*V1!{X1ba6|pJ2p|D2u2tM*%Ku{}
zSj0a2ZE`NXWtStPr{}OpomNlsOZhHI&L4Zg6wX=6;HXDM#W4VUBD@)p1on`${;<b9
z>OP0E31WQ@zBVS{bu*$};|7*Q{+dz-Lu;LLnWW)icU~LfAZwh)S7M)$Qcx&-G6NDY
zRRnR0x_9)Yen+Z5tV@&i_m<dx9-wNUA(UImfQz=qy-}ZQtiAp|;MUuCl#I%Dd~dSW
zcSkP*FVq)XH9R6A$}6C!2(+u~{Vym9pXTXS4V<$Z-bq}K)#(cRGRSi4Ihal#$id!W
zC4K6E9N%>QCfkQ<?;lbA509FbCi7M4<<xEmp0hQK-%9X*%M;`?FBs`ZOWF@|T3#BG
z@-KLd<EI7j7t@?fK12FjS5<vF;_&RGkSIwbKlh93qWpsF{^uUpo1;1T7$-6&n_&qA
z;q=qcceHGeNZCV4?eb{|{`;2^`0-3M0*muLgCwAQshq`VbVMWWeWh+}-b;*)*!gf5
zg{wxabvd!KR4eembh$7bmwc5H8%4uc)B7f4<-02VoBQ}fBqXZ%S?iFIr*-hjoP!AO
zJat&?v&zGSxPFz_FC(_hFBT#5D&iOE4;KY(A}k+^ZDyz#T;47G&T<cA;5O)y1jOY#
zQepW%7PUf4NulQgjupPIyvY|-);4_dE>58)<`S}{x$hYn+v-5Stdqg@FTx}8JWYFS
zXJ!$-I%q)RfXIFmbcO5HYvW0T->WlqH_~}pK2;;|X6v4}Ra8k*9_>9%5<9R9HAoHM
zffMB^2F8;vCK;zwvM^v<;NzXVQo5Lxz1&M9MxQ=Qzm7rmmDIXr1@u%hn^N#qEx25-
zQ!u1AapDvg&<N>R-{x<zm7s>Ipm{zN_HtRz_EY8dX`3*cGq<n2)v#&cKPO5;is99W
z^PE^0*L$$BO)u=y_z_*z;b4~DQ0xw8Og*7_UKCb%X{Le#5^(-ryr4_ZkW%Yod;6HK
z#;N6vZ|29mGVNz2Mn*h_d1SszhzursQK}t#KUra*ql87{s01!5g4+fv?XGm4w7Rrq
zJ%!xIiS+5C7?j)80%OZUE;(-{5t8vvKl)+>;usy9zrs1w2Z}Nj2>plNMB}e-Q@a=w
zyyh~ZW1BTly-AeD?$pKHFKTaoI^;cN%-CD!xKKX){hr{mDpg)9-(5N5{ue^t5>owS
zy&7norl`%@`-{@CWPgST{Pm~j(}gvqOThz{DE8<vI;SEtt=saNLYyU4?$?f=J({<^
zZW629Hu6K1MSx2Y2zK1c{_xuK%gdOb1L16?!t;jai;fk!ogamA`wp+XbTJn<ddVRb
zJkJm|mXRhSzr0?E7{B~M>;J{SokzK`sK+K3;D7LPnJ>u1HrjhFF|}JOwHwd(p}$bz
zs>e=j-|<UvVc*csnOn{U<ik`ci5T#?BrNh4FEANO^MA6~kC^4JVZ^<5SSgGI;|V}L
zv~pb%iDx?Dk6NG2#rb}rvE6v7rXd|}QW01SXg?Lr?9hY;FJ+P;W|ewh3+Fuq-IiC8
z1hNy-WL1oynOR>(Lx-}4*6GUY4t;g`TWhCplqSeVT1Sap?f;xccPLR`(MTd=YeW+R
zwh;5@;E_esXZ}Zxp@VHosX#+1h%-a+^~}FwS?c+;|ER|*><)!Ht?*Ugk3gSl67QT<
zp)lN6rGM*y4?#jzie8C_AzH?Z{T7ZwI<c6SSR+ywHeg|7TTHBsk?F(D`%V%ZcAQ&-
z281yjIR_NbPh_te8eI4!fV}_-6eV$?v*&rL1_cLH$~~rTO74W5f-vDsGe#I7`H%j!
zqO;4E4aY1FYvDn<UkBh*gbzJC6vG!nZo5@pChpSXdY#hxrJvF*Am+JIIMUG!MODa9
z{M74t3pn;=2%@9FHuK6(zNCa2Ld5VaIrWS<GKSG<l2P}EP8#zn0AWYcs{(gd_F#Ly
zhw?tsv@s$M^LS}Vo#=Mb=QE)wyB=DR>5Z>{@1PN}^451eY<#F10^T3!PR*nE6)Jy>
zd=h04I1L+WcAb2LELpby>GzMGtmq>vO3FL@p!b(vthUFwA@>@wTxEF6=)fODylp%D
zU_>`eDhkb?MPcO*zpm6#I9c#lo~kiC@6G?#&?fdNC-yy$$+ieKR)~DcIy%&u30I^A
z7?Yct2`$3uoZxXjGlw9f+*5WZ#5@Kj;f>Je`cj%jo<i|*A-8$8IyvC189$XiRSdUQ
zG_jweF0YZyN86GLyW^IjO%%NJfAAQ8w@ydAI*6TT^*9Zc)W2xMK?u7(;*PkYrG@JT
zi~1M$*{^A<43Wl^5m++(Aqmud<5&2Ck%OAf=A|<vloj25mtV9-93${O#<M!knlJ0w
z>gvQpEsVdAcZ7aZ3m7IXA)n_$VaYu16vB(3MUL)2ZpII6cF?%Ljee6_fizInb@1_6
z`5lp-3imh`3zaE0k|UCo-+fICI_{`Y;AsusWU{&u!EaAp=dXv!>iK*nt@7?~W2*{5
z_3m0@-`h^5^u!?^@DuSV()%<L6LlfL+hu%N!Yw@YgAG+4vn@X)Nk#1|xFLbC>#Ru2
zUD_aes1PL)!+|2_&&|2;Ws;=?V8D5^)&|ssxSW)lal=EI&zjL+5%`sJ6chEay74Q>
zqRDDv)<a+hClT<RC*TymE)D7Hks&X;>Zvd_s#B%Y$SH|R|L^D+vhtswx)bh{@gC{%
z0tSCPU=vux0y>wA;Fv-;!<HWTY6)`D{&dPlD@NXe5@(R~{Jo2kJydd3!iKct(;vJC
zc<lJVuUo@kb6XmR{qNB}KEKK+%tz}`L{~CKBc3^YUspJ;Ma3Yuo-*mM>C#Mx--ONz
z7BbOqy9WyOi5oRi?LR1(jxBF!W?b`krzRB_h?4w9`#30MC31+!DB-I5xjX>(K-1`J
zQxlv0Z4{y^V&fDN^H5h2L!rY;fjdPMFEir8&b~4*f~wA%pDXidvmbdbyE^cqNxUc;
z52ysj;BCYz*QQM7*tkdUjh_rie9xMlD4kEBYSd6@&=<Pbbx%_Jw*7qm!vO|tD6z~V
z{y7JVfatx9#!*{bz}3v%)S$_W`-H-N!HV{TKP-$lf!v~W=}EW+#e(1!FYXq$kqnWf
zxO~rUpNHzkU!*l_YEN}Rz>Nr>(z=nv6nHiDg|6hO3)@5W0|VR@mdb87M+!t0&KnjY
zrjru07*6$D(kQ1BL3qwPmQ9>7p(t^*ASvRzs<Q2*^?bA?WVvQCGXe<VtzbXy&<8?a
z-@D%hPOUU)z{R>7)g*LH!1apt`H&^6z3>(=Y*2l>X5Kqh%FD}J#pZ$*8b_MqiUBfa
zZEY*fxd(YvQJG6Wyhk)Vka;hmN`0D(jYp9Qy-LV4E{Zds$y_RZe<FS!Ht(6`ejbFA
zGSECL3QH4RzB@O!_vXv~>o>~Q-rM#0E0Mj|z4%79*!?bEQ;xVfSzDh(mzPOO*r52J
zmXG;Q<6>3w#OJ(WZmVWoERPw}74Y@%>WZe`*HXz*XtBtnLasuh;~Lze#X~YrH5R~0
zJ7j8)lHIh8zP4anTZw7rNju)kwnBdTS=AY(Bv&c+iREE4^P2g(ddg4zIFr_?&fEp5
z|8#;`@b2oEurl&ds2ebIn|(T5e{rbF<gb7(clv{zcMJ25Li6wfbIB)>4y<CDe-DmN
zMgC8l_%bND-}&SD1=gPMDnko_Wpt`ib)7T0H=U22J9WtuqBufy`S$EBILxTRZBQdI
zk(y9Lty!PSB_T_%ePb1|L`S-{W|4s&4aLmpKg8JQ?|<LqacrUR)g5){{VL%%I07#y
z1ctCctb|{CCgRtFj5$BP??bH1edi*rm7v)GYYU&`1W{KP$sgg3<t?Cg=Csl`2OkJa
zT|$%Mi~%gnEti6?Aue|c90)-VUj5WM@Vz$|jeDZ-aRNp(1SPvW^X|v!_n+!%$Ak|Z
zLjlv_wNag$x-_Zj3_+93_kk=n2zJeTQDDVjjn$}!Tl2Bsi{0>8qJo!*=6Nign2~U?
z{zv?y@Jg!3N62UFd#^k%pKa>Zdgf}3T=+3OuweM3XVU26KAR3NoI8f)CrQo-`dp}|
zE5ouZvVYBttb>@l@nKHntY`-s=kU-&?>L+)FpZ;Ifv!n(@1n&+_M>Hylo<=?M_BzI
zgi9_Tia_#Dtc=r7B^8}IAaUzzWGpcqgH9c1!{}2c{rqg^C6lV%(BJj=)EuYWy|oeo
zVZGA1^vyGNR<*twiC2_R;D3Ce2&#+RkhG@v<VKc1pf=fxU0f>^Q}-_@c;y?9=#l+W
zG52C=Grs7_`F|RSTyNRun++E*-CjzWJuU=R=M=twCzX|qAkKSg=**e$rOKHA|5S;j
zy`~$>YSijfy_ew5|9hCE+e3!-Xf$ICh0VttID(A}tOTN~tdLiN>p6ETM?La;&6f3H
zCz&mK*jm%pfU-SkAsNCua_ZNTl3m)e6j87^-@ov4G+3K?)>nH&c1X7fX>)~2`va-I
zyNK9REhjL~412yRV%U>WU-HB4vw$l~7h6a+5%FCnZ;_`br*HJ_sh=(vYiPy!1b=?X
zVvkI(lujoUQ4tP~Bi?=X*@iIS4;oy8Gfa8!{}y*xhO#mDm3J-)f%g$_WE3S%tzVO1
zJ^mvI60CiHm61BvU6_{sE-)P%Qg#EN2v?C<Y7%J8gyee0#+!2nadNI>cGR<P^}(Ta
z-E#4Fmx{uAeT}1=i%f>jgXv|$))t=D{q<jR3@;f5NUiM0ipMQHVZq9r41RUs{>5c6
z+O@*DH|2Q~wG_!waL6_Lxh(Aw{~~d@T;x^5_kMHMi`%T8R0ST2bAlx)<KN2906D-Q
z&~=jJ$?DCXq<L!pfogZSCvu3!GZL#qiiy*5T<uvLVM`iO(}VE8LoR7;P#3feD1WW_
zDjh1l@_Q!Bjp03KuIhg}IKRWMGA%pCmrij%MM>#*fmtmRgCGlc_}Y!HBm(pnnd^L<
z5dMy35MIhy!RGp+8ArB-kLp9Z`NI5&2O%PZ9hZgNAT%xjg*&q^BmAB({p#ymTEBf8
zap*YBpB3*~$O)KQx|x9~Nd4<W??Gfx4U#aE|3DY}`C-MaB604pb;hDQPArHu8Yn40
z2e98N(QVi`f9^fr23R?95CouMzurkI?C-Ca77yruTqu_k1`aufemV^26Cn-p+E>(2
zU<^Qc#l7i$Iyv$qqetXi9hqKc!E(jJthMh7-$wBRq}VVH4sL8z3FnM<2-f2C2F9Gf
zv}Ty3;qr+JpZq)rxH?wUzi3@?_c@S~7LXj6Cwj*ZCMzk~zcODpEPEt;y;pljkb$5C
zo_tQ-s1TBIJe&6d4P^dduPxrVjd9Ie{KR#MPOn??(F5Xb-SO_W&D?Nn5_6dXG5HsK
zXJ6dh^b@?USs&ngX!+`@&3&uw9|5Jo)B-YQd{(KFPqmA{JPQg-m+EdR7o=Xmp859I
zStxpbMJrtO*8)AR{wgoi3!XzzXo3(idi(0)x!44YKg{Tn#Lvvp{~M>2R@t5W{G-4b
z`i9RH<=}^yldeX_@csL4n&mH(ULwW>&}xW69m#O29u|#P2nmBOf$+7VZRA+Li)1F^
zWMH}Vsm|KnW56UPn`e=&+0Hlr9*~_8LATU}DkYET!71vdgs6{-XKo2!vrt}LteWGD
z9~9UN|2N6erCz*v8*)i@JXO@Ii4Ld685h9E0a{2f755IOItCb*{<5j5KiVsaL-0D?
zv_qMZoP-Aw&on{A)-?IP`CTx#Z)-z($B`)jV1F|`K6t7Z)9l2%D4v`%?v@q(zxjsn
z<BQ7E@<>qiDCo&09L1ikL5y6CFEE>B@>k+-F!VdWURd`DDd|0`=y9Lwu>D`06A2-P
zg~ygr1om7_IPa|5+{v)M%;S?U#@CykDNxs(t<5s+BvYsv0bb%m;0R$hlN2ZAxrRgx
zXAq|-2yLO_Qq|oTTyq>e+&)Uqi4HxM=iL25m;4B^;iw4y#b6Eq(x<|%7D+o<jRhW!
z@oeks#y@%PAx_qkC6^|;5<_t{;mVxxBN3RRd)_oiOgwEQL5>uEh<!ndF)D6Di1r*1
z#ZiPOZ$pYkSew^1WG@@<OS48tf5#e3d+o)ID2q`Ez{hw$Fk^rmzeU5w>sU5;*>-MG
zi-z&4mB}X)CeEi9_6U_}w|XvLUSIljR_{`)YLskULpIfn8)GPo4o?lvn-327t{msu
z0%jIV7<uY>A9X<cS;@}oA2(>12kBxrUKIB5tRf}QxF6CT{yz?9KWx~p5ou1+usj|-
z;4<Z`;W^jDMYW6I95;4-!(7ZKbpQQbi6)O1W$eZu9OxG2R-P1AD1|L=$+07$am64C
zMAAC|xoq?0_WVo;;Qj6{_?ActxyYchKQVsoqA$mU&`VxOFVPrUFFH-@EN*Cm!nSeK
z9CAf?)4{A}p>UqTx!HLol>3R0=rg&$*X3qLc)iE8jMH;VRXWe9zGuBYU-CpGi|2xC
zS!M-dTJ<y}jXhaJ+#irM5#BU2Xg6({c`-_fZ8O;>sZR0gNZ7A?Uwq;5cDYWqgK-l%
z&4jsm&tzvM29OKvp2Qb7O;v2n^(Rse<gymuu;+jj{&W*@PJ=sP3zHvy37)ZbHGHXI
zJb0zl$NMd||2o6Az$B**H~qa0c=-|2hQK^E9$(9wDAsq{S2z#mV~uwqD+k?oL_Y18
z2D_?SbXSnxx4nH>1(EpE0z@qE&M*NM@?r-LOR~gbB+iRAO%4}-2rm(9m!_1VE4zlz
z8_B7_-VDjfB9A7A6rbl#%lU}E#t7?!(3kX33yLXEK<Lxob;Sd@uiuj!<2wV-f>5js
z3X!BTbX<ocS_XixTto4Ft~f)`KMEo5aUG}KN>wq{rLHo?Jw^~7?bcPQbZ)pHGAUDH
zFu+{<JSd+b4!sCqp~bt7+#}A2a0#yPf$d(`3P8elfqcJDu8pQnosF}bWAoky8i{!{
zMTynP=JxI=?jo{K;3$aViPhpYo2+Y}YAC?94HiBOu*L?#{HP7d`1^kp9KB)1dlORM
zB7v7=;v_LaahfIu5L?<^c}U>0`ru34UriqB#Iy0{ij5~YYLIBQS?irw7s}r}N!@v?
zLgsNEFg%~E`s9YNQ))}4#GS?I;;oR70DeYHtM9V7u6a!2%C)%=7Jc}@!SNUs&OAw}
z-sP-y&Rf+`^h`dzz852$E=+lZ0lZsu;P^!=CiVL-5qXI$WH%|n<&?-Kv(&|l1>^DO
zxiX~RaL~tU+sv|^M#5Q(wQv-?V}Yad5C$7kvV*(HVh2My!eA-j&ed28QhYsqmHDOR
z4Ca0h!<?IZ!w3XAVuKcmzpsnHJX%$!(_p~;V*DF?bAS{l`9@ss8SymnlS;%+D%$_^
z?Go@72@bjL!S0TK&=>H0FR9k#t%Dc$#N0rr$MjXNcr-4;=cqlW-f$;nLKku~p>zSz
z1MJDRnR0p^m`>?^d~-zJM`;&X1e;=`wXqbK(WOu^*mV(D4H%3pFghtF(Gr$OaGZG3
z8TH>?N8rC&sOU5NN}f}ABTR~Un=DJ2120ph)XZuavu}xKhdkf>z-|x+d`V6=1Zgh|
zf7*KgEK}^UlK%Z^l|`Rywo{+h|8xcgJ^9h1K13k)G@mQAVjGlT(@%)Tck=`bYD23i
zc>5E_T{c-VKh0?}xpjfzz)@@WXG-qRGIwSnx?h5J{RSGO=oOdSmIWd2CDIOy7V_u%
zWM1dEG?br3{@XmoiA@GT%7~=+SDy08)<SumdcsJYR)P+a@HGF)&N$uh1Bv+HD;~Ui
z_VVw%X0=CrE#cFnq$fM4juL+msKr}Ox~a}qRZ9haDX!eW>1t~!-?T41@MrH1G=9_3
zJ8GSSTluA)<o+<v&xj_FQr{y2q!^58*A>)1{Uqe?9LN3B`m-H+{I~tZSjyVHHSgy!
z{kLn|(t<0OOro8^9+9$F4~-KO<f2{<8qQ=!e8hS&=fzig2moRAOI&Z3&7K!?A2OqH
z7X%a(0kg{#Y0^fxHxt>5Qf1?zgizFj){)bZvd>j&<OH!D`$tP#{RnW(kCmZB5U}%F
zroW;rJyy99-~GBK#)hvrKVPN%6>v_rFFwq{m|1M9XgvOx@ncJp|2D^QQPGr`%c~LL
zDXJTff2y__QsNxNM%)feYTV1t09p(q1|S906oFZWDo9{D^1LXiawMj0I6<l|qKvMp
zOsQRMM}17<dSoy9OOs`EDU%c&>fPHogTzJo&k^2~|7anVOptXmc%f>Ii$>TTE8^<K
z&pdjoRH|t{N{Lvzjh7{bU&+V=-MUE^gz8TF#OmGV%1tpBIf&Jm^2i@fE4w6<ws^z9
z@71&7JFXsrSRkPPECQ?zYeTA(*w}|<T)E0$CM1Ll-8AfC>?=qvr}1LSZjEiDUXvqn
zi23s<K;P2EOWS&B8QQx+?7f839t@_yxm_5zxum<C$K1Crd+KK^=eGqwOvYQmoJ5=l
zxubvPRj1mVYt$3S5o?dRMC{(oymPlcb9?P@I5i!;t<jyj==keEIA;D#ttPfWr3sBQ
zC7BmQ0j4<9`owLYU%8T}#xv`xHq<<2kF(XQHHBW<2FTX0cwfG7&*$rHZg8_;LIme3
zv(N8Oe$lu_juk%WPS$IT1|rOgQ_06%h_%2s(tj2;NE@=z%4rS{(o^f!O{kz@w?D~p
zm$QKEo&8mCT&%bc+gjyer^0)?80Qa^`gVs-7g8r1Tp1ztdGk48_Ylp-MoWzv6FdV3
zNC?;}g7I#HpuL^;3t*F_Z*u!lQpmklge1bVe3}Q5)LX+Db{5LOlj(qCJ)Acq5{sK2
ztFJgD-eNIulyx9=)bQuASZ^Pom#758ne~yCJV5+<9vtmc=nqDK?j(yj<uzg%_&h_R
zUP106Ej?aDHH-y?wnlnSWvO*8Q<caRmE6Dlf55RrcMA7&OTqIY>Dyw)cE4Urc%`<2
zpL@AgL7X@+%`r<nBXCC%WMsAl|DUq}_c6t<dG>M(z<lTpI<+iPTpH0La7paG`?b-2
zpP&Mf?d3Y8-4lxp1R@DBFO0&WuM5e;uNw?*8#`DU57FM<ne{*n-{c-vMxcMheZr?Y
zpI1oi656gJL1IXIEz#iogN}hGpfGKFX%SUtS=D4Z{>Es%@VP}%T8_h{d)uP!=+N6s
zT5|ocdLq1;kXT9`Cfwb@aoaUVu3Ey4h7RvpGDKD08_(?h&+rMfSmu)#s^;N?mdq)L
zLeg1%945<-I{mK_c9Vv%QCwHNLW)g_?pC?o^+*LLZy{Mk3y82msSucGec(^39@-Z*
zQ)?8gXbHdPmCwO<U1&g~P&8X|w6{(}?X+6+dB-CNQjgq#`JkS_5Y7XxbfzG$xgDWf
z?w*hKs{SBK>fLc*i(pk(3icF8;5;ahsi2>R-(fwUaR0i*9r%zX6|eGp^SizNbCO<y
z9*+k?vevg;w2d=a;Q$8DP6j1K{zN#C5_F**j-o8t%B^CzC0T~0hY|6ojJzJXeY5QS
zlGFR`M#bb$2<papGb6G0T$%96;N`?r2q({tIztFyL@6n$^-iSv-kn=rJAx0QoLNY5
zzyT{mjldHeBbv2A*|Q@PEnKqeXP9z^;qaw%58?B?hZurY%9KRa_-T02FA$>$u0P;&
zT-cv61b4%^dcDu!qksStR2aAFRaH+SY@V3-cE_<{<Z}rE-Viwv^D<4ez&cR{EU7{L
ziXKw#YhK-&kSLMc^Xs_D&;VO}si#tr-|IwD#s9<{?g&?0IS+IK&+RutQxEZ{kOawN
zAFoHPFAg$I?MLsf1JeK0OiD|fqQvP0-mFf_HEt>_@lUy<uiG<^PKsTkuUF@<dt0j<
z>`Z|Jh5wAG_AY}GDhb=!cI~0*y;#~UIafPr{!B%NH;d;RqiT@WvNH`|(Zqo!Kw?hy
z+QP=MdcaMTLW6QhvEKf4fu?g1<NK6;4f`UHa;K@hF^k4$Oym>_%bL=T0p6kl?L5G~
z3D8My^2vOb=@UPnZ~t2)cirRJWRPOdEqzu!ANJ|)3I>lW!k>O<G3rsCv_M)&DH;R1
zIbCuamb3jP9}idCs|9?&(=0t-U*P|2K_)OPtW}ZW9PZc$FG~guV@rU8XWpylkHouw
zDzML&Y<#|@s?=3h+t)|bLE8NlA&>-=6r5fQr#Z>Zvzno>j!c$m<`yOyE$yH?U*+Kc
z;pw{rss6wJpZB`fwf87ol|3>Oi7T^G*|L=xvLYk#GAbF7J#)(zk-g$pWRF6K5VH5)
z_xIBK^Zot*d_K>4p67XvbDocLPGiSmV{E*KvABCWzD>saYl9NgQ~0g=Z)`jkB+M9p
ztjN8PSReWsiJ_P(u5Vl?JpO$D<c<sC&&B;qW8vG>TfQ&k;Im*F>qCjeklIbs+ypxY
z@41=$NxB%$szD2=iRn8EY!o#kB1fwqtFwXWuBCKV)0L7%j)iJ??-BhKCxgy1mq8zH
z8MgL~Zf$c=mcfk-5*i!1lyGG~+?fO}e^qUabR~WKNtwg&f4oI!!k=JVl5YyhBI()Q
z+ZikwH;;~5D1KDTE5CB?(v@;#3AGSPJ<dEkVkFt&#Yjk?7zICRyA>?n2xY9@V!ZU@
z*b5Q$BzFc5sJpD(qzNpK-G8E?x>rMuDPU9`3HznAsDlB2uLcNl!`;^@d!PmzZ?^Yb
z9u%qy07FG~YNAvlhsY=WjL2xjUokQW`9pi+lH}u3^7nShwc<#vijT_kVP)UKgYhSH
z>AJExxy@G#!dSIP;=aPGAjJq2>ZZ+&921k-L{TI1J8l+;YTsX>6sw=}Nu9M#`2W)%
zb}1&X60F86nVjRlKb;-eq5uP9tIAAd`3+)+CC_i!IbKP3rTB#W?>xc<X(Eg}SKsQe
z6=~NttgRTD{VAb&1EkKvVd>a&py=#X*@drROY^h^%|gr>B~NU%ZU=)rF&M7@LF*G5
ze{gp|-i*T+6h8RKi^T401lN&cZ*f^Vz7eMAm=&DnDKv_Fvu20-D@_KWOYcow>hKG`
zzM7c7z=Ww~UR?|`KkjR0fKsSIl<`98tY6mW-;CI@`&Y*lGVQO8BpD(~o_x!oi5tWK
z!6wHDbg%S`drGh0*SBp`$G7&Obz6iqHDKEyyhlS>naNHE&*mNN9^d=@Gr);!D;Eg^
zSRnMmr){b-(~M}&!mL=R6vQPiR5C)Bo=QDWHlZx!yeqUA#+|Zca%Uyxt5XUMs}hWV
zJR(78z4fVw#BmRFqlZFQ$b~|n1;wRS?r=>~{LpnG%5A*%uuW!UlqCNA{Q3;DX;WX1
zsfV1+$Rf!Z#rYl-3~K~Tzepp(U+cUN%2M;Vhf4!`jYpCK@@6+uS<PH-F1sarSz0zr
zCSI=mak3t^THpA=NF1F;^4AmQHx><%r;3rxdr)WY4+b44Bei0UW-`TdjK2z+$7lI2
ztX<+=nhcJb%9TgiEobb`g1?qz5Xse?h`fXES5er3uG?f#TStI2frbe>BY>@QjxsQA
zO@fUuM0+qTpsLkmYy0caMo>L}`{M<t`6h-=$1f+E=sRJNN&cEIoGo;K>4D*=b>^re
z$$xKzU``}%g3!(vcg4<g-tT6Hgx9e_-SDEKE>L)P588Gguha@^)%~({93lP8W}@SJ
zz!^4|js9`YIM`SgjNl;B=qb@<EVpB#o^NBC_(dd^8K6Wt8wXkOYuC-Sl3f%4Cn4va
zkYa&`)`uW4hjFO*hKx>D&*j%h)<20bGvB4!xA2fLy}%yHybxJ3z*wod%swRVB;!4q
z0pi|czzr#8sQakcX!7Fkv;6d+ZY|22QfAM;etVhCzCb+U{C4+fmdjaoS7P{;2+@B<
zVR@(yCVv=>4dzA=vQ&|lErc)`Z$loCuv%0_2M*y|*{1h@5Arxx9$;j$KzieA;EL7A
zg;%R(Mtg+w?9UkOe#Q-r>Y0o>_M!O!;uOQdquDst(XAt>e;6%z4%-nhUV#)711hE)
z)kfH=djoariaAI#r5bW1?z!d6%ba7K7FJy;Ce?LeUE&&>$GLbLKF6pT!eX;jSMR8n
z)#xjNVlQbm1RHx;njOh|=Rv)lR-IzF^MkJv(TVOo+fA~>eG%{}4bWP!k->(AEjFCG
zA!0%?=mv3gj-s75=Sk?OvV7nnW#<S5#O5|0WW1`?B7_5U{CrBFe~u05(6EFHLVCT6
zB3I-CvFiaY2-mTM0~ySqNhl+NAA4%X0_g*C?2Avc1sjEWTyzSuOi!L}b15R7T=zYg
z{ga%>XLQWnT}E+XQW(H}ia^+!V?@Td=o{##U5MpFi1PaRHe&2L2)0Pee&W^u`sRn%
zkH$C47LTMg5&s#!SelbTZRtp_!`vPP;)P}J<3{8cudU5lr+teWGK?2Dlk9yZ$|lR-
z*Dh9PG~W@r4b?plfUk>?)6f3i%;l0LcHUm=a+AlH&Z>(PuFk(td?G2IDU|&s<ktd$
zGlh59E)j29MOs&?=`SG2{+rTOV+v@QWw+EQ0)|r-*Br`;pA)n0Bp4vIeiMtGD7@o-
zcbl{f>$#aqMhiz<MPY!i6f-!Z;~X2Du_-5GuZGZi$k#-m_U(?ibbe5wE|Z?8_Uf@R
zxUZ!U1oFN=X4V>i`t(=4rw`dIBT%Kz<Q`2>xhce;T>K{Rgd2#TN~xSy{+!;AAC8#^
z$Bk2D&^w|v=OSR`3lc82c2!?8Z1a2D^O-;kN6^5;1zY8gRW6P$6LN$f<o@+YC2H?T
zG^e~256H|c+oK{|D`$=;W#I?}+CbQifF=IUKp%iO+_2X9JiXwPTqOB%Lwh{xA~_W6
zCsaMaTGH_)Ppj7c{CnauBTw%zI#7DJ|Kp!1vEK;(;hk5Fo)lxcvxVOKh}xp&HW+7y
zo@Pil4M1vvNxYwHvOY0eo6!9o9~2ldut(tu|5jWoTK??VHOva7eC~nPeoTdxr~!)q
zBdf3~K}o{PnIp0F2S5FI>Jd9*%lDb1KX%aoPAzZ>p`b$bJeP379d*X@>h?KB$s9Pc
zKogrCVS&3v#U&y*=>>4XMWy!+Lu%0%+LJrWHc~odaBMgk7IyaFkh^VpNtVNIa(DWi
z7|U9?IR)ENRvwz^9Npsvsg!+hJk2Fu5BNsV0pE)hEL>vw-7rn0mS_;L$Ik>j9nrUO
zzMQU}`f{2u@RCpBjlwzl0px@z?AN~3%#XnE=4$7!yhltB{;fT`YfW*^_G3mU;ZM%r
z)Yk`m-=B!Do(DvG-3=PHhtmU74KQ4uOlcqH3bFV)&(HbW^1N-@izKoOYg^YxDy;B^
zh@ls@-AXCk*0~((cg{?QO9~B|F@gVSKf;XEB<vNkx4bZg8k%*y0pgex#zIa1Q6
zQ#Ah+5{Vp{Fi`HZv!+pMLEzLI#U96H7_P47q-<fYjL0*MjAcVbQzKgRqs*a}X1O2A
zeQ1+X=>@xuB{UO1UJ`;GZ`fyFi0*CTJs6?qCZRtSg0CYcgrX@{Tj`eF60NOVQ81=e
zuZgM*PJ+_|+LdVmM``g*wWy9VGOQWb#MyLxYT%T?AL_Wm`>tY&N8ao_NsZ&kZ#HB{
z*7syiYJP~cT?xMFbb5l9MFqtl#mDSYBFHB@J8b&v<n(}lWU<M`bf#3|F;!mrX;QSH
z7r$V!d`64Dgyj_sO<XXn-==oG(nCB>PZx;WGTX^6WzZVline*$SngWUw-(@DITeSD
zjs}TW9$<T(l2nj@qt!rncuvGDuOHSi3)}qSCn|R6BF4?A3S(M|&S>>rOo=8?>~?ZB
z@;W?p<));NtO=(f`;SSE%9m%Za-#i9o@w^@RPqea%^I2N&pV8#S$!kK{;KCp59|5T
z-oz9|`aUyUqMANOagkzya;=f(R46<LSE++O-<I%6t*X0+wEXd2+~SFTc*t!YDdDuf
z{_8+NZ`oQH0~AitJ>_=~!E(mSc53jXgVR~u(ZS!>P?iL9xc9}#>Cck2UP79F^yMWq
zc|;f=`c0>s4;J`oal&WEpUUF&PH7Zw9<;W<<{7ie|I!na;`aXXnaGz_6oWQGNxD4C
zjO6+V<U}yMtaxy1MDYqTPUANgG<j%CQf*12u*PqH7ZcrRjA)bE$hmIcvBDjIJRO=9
z_mBnKG?7e=9gDXA9C$f@o)uE%SQQ1|Ojz`{2$S=)&)->#)wM0L1(2M!PqGZcL}&J`
zs81f#mN&P?aAq4;SGKX@J?@3SMPfXjHwN~)y@+BMG?Vf0EUi``gZ{~gp*)L0i83_h
zCjC+B%%@9N!RjZiLt(+ICY_HTAuJ2W9^Ty>i#a;U=A#(n*7vN8Ai1)TqVU&?6h<~d
zbib}cFM-OP1!<#Lr3p;>#yuxq@JIj(3npcL^LMv`4=4F)y1v|41P3_zNtHlhZ`hr^
z)HkntI+)*mVvT%Nk&Lp&fCY~38?&>Mgm2;N2SemeD;+x1nIj30Oh2+j|5H;*DYVe$
zaDrwsdG-NQ0y$)Fda#$hS&((Vt@Pz4EqHXL;#;v9+5;i1v`>n(d`ZHN39aHoLEG6h
zfB12N$qPnKOo-ZJR!HK`%`4=KwbWLVq17IXc1Pb%0Z(Gjx4&kwx<<ZYp8eVyMwfMo
z7BZm2*-Zqz54T@Ie?qeAQK4NS7<`0h{dPz9!G^!vqMcgq^x2PAgE!(e>7w8?fhnS>
z{=&?@4vhv>PStl}dvyC%T43VwH=uj!v)_g>CDLMiHjBn;hr($70e{;2WG8Ms7BM$P
z+8x)+Y+@EnxTAFv2!_7wJP?#7Xm5gW?d*QF&TDqIdjZ8L+1!DNb8=Jj#Fhfr`_hEs
z*^#b6@K*$$HNho8RrfK{a)79{ac+1Ei0a<G7;7qY@jfVDOc&w3VUSuPApEHSSJ1?J
zXaDLy$wc|D{(Ww26;oY!SI1CCh&9h$S=G|4$r{}&o>)O1`vd)5!ILY$?Y#eg3}ucl
z$$x!ecwaC+L2^s*vF%oAhVaN0x39Fl)vQ`}GFf9AH)G=~;Dhw?|3*(O_lmeA-u^eq
zb@Eo=P^YK~R&E%xR>Fz(x`^_(sN%O?X3Nso;Kw7{6=@+7v6MF=0sQ@iylKP>-Gmzt
zonpPDR)k5D&erjktrHf>uvz^<>9^yuD}38mNnwdX%wlY*vEA#V*Rt6i7#avJt4{ZL
zKaw}d498Ck;jOn6HNm$z95%m4exa0IVGa&T=On{2c=$?E`1a-6%AdSmY8EP7Ry!`B
zh7$eP8IC>F4zNs#r8gem&T~wn6y$BNA+;2E{4kvZ1-1fCWsD=Id@RAjgC1}3S{b~E
zpT02ur7HOWfmqT}IMn+Z3nDzNO%%#y?jwf^)T)@E3McK<K%f_V+LPF~tNY5ksr(l&
zcOwLdF{7#)S2etzs(5X?H^}TCvvvPuzQYZe1HU;i=r_yUSp2(iGz~|TE73x{bH9jM
z%jzA7{cC!o^1>Eqmw7Z6$9`8noByGt^-$P?xq(j$GB0N9>U~QJ5NoF&LA_-0Q3)3w
z>%KXwJEZeU-+<evYrED`9*&HX%<)hQXtJVgdzRgHqVqQHo0C8zKI-bLpGBG`tus`4
zHz44RlSd)!pViuhi-1Cb{$`s?L2?y$OBz+4?99JiS?<z>meWL4{2z0s1&F!6xJW9i
z?lb&zcrgt9*YUkZqlIO$#KX6=&{HJ)qh-83eG<j#U00;lOxt-ZT*Dx+(e$nD)78gZ
z^GL@!`6$;OgP%EIbwq;udqmFr;fEaa-?Pa?zj)n$yo5iaze~~(KQ%s2k81l=S4<X@
z-!AoB6EOQ3mk@M2B~wuUke^sdJ$cj7*CuWKN&MMR{n4O}*;)v3)x+b>2|^VQ<yM0y
zIcLOsuUVy-b-|Z&@d|&xi&;+}cIrRWiO%Cw^tk{Ed@1!oGVrT~))ypFP&ub?QtTCb
z3+n8`&Qi7p%*Lcz>AM9c{POK5u7nY0x?l1}-?KN`Qsq~^AH=H`d#&rd{`5$XG?`eD
z(i0WTXa>DTV+2ws2zMg77h7io<W5gvsgRY5c$S#g?AzzQ)#oZui@8el-;akp&-E(M
zPJC(>2o@<-=y<jw%X2p*Fa$YaC;RuN_R?kYJZ8xKgBuMY<ll;fF{f{^Gc$%1_CRD+
z@)W$X=X6~bv3L`HL<03K(z3HcY(1=88wzGcFV$Z)f1Px?1Y;1%ko!3JOV&P&1P(K&
z8W@<8^VMqCe);S`8HL6z9!2wQu%%t7)P1tNp%C&K6jC$KN(O2GDJ#G=N3?f(6xx&D
zD$MaP-z`}nf1dF@RE><-`_XGd)tza+?qv1!4`y_qs~Djw?sV{H`y7cL#i_m~xNc}e
zODbDoTES4~E6T*iYS{uUM#xLMM{ugwgV|_l@qEnnnAEb?K|bwsZkpEj+T6VFrd6;@
zu)Tn{3j8F%NoPfQidt&ZeTYY61AEX$g4lJ!$R#C>sR550Cfy@Reuq&xVh>oBW6Y&V
z{A%yMu0z8P3s&Tsk`Ju<tQ&*Q66(iOZH7<>R@?-k`8-en;XPs(+J0<YhPwTHv_oFV
zR79j$f{@Z?XcURK`fC^92?%}`CBUw{rI*dSX(XSrltjX0JaOX%PC<a-0phM9deuKZ
zN%O)bKKYAIKdpFQ=R}7`;yc(W9Jb~VlztntM5mZuM-66s`r@cx1J+p&D*#!09;%CA
zNG!%m>Rg+C$4v{9k)EtXmlu9Z<?W^P_-f-{&lAe>=MfQpYjAg1=N`4@j2#mEr>k;b
zbAPYVFgO<w+BqzW^oT(EEzxU@6WP<QuZbLFS2Ae%uxXzRjWe@%-+fn?Ad+-Xtyl3H
z_S)&Q*dtIAgen$%HWN%q`TRxXEwcb(s1`V{qC!!Sf|K^|wC1mlN9$o>=R6U(_jOci
zsos)XgCV1d6vhBu^Z7q1)I!oEbE}st37=bAh!Vqf$iwaw47^FYj<}mKqmK|YKsv>G
z8m}qJZV43>%9CaF)LOs0S|t9L%`7i@vmKw(R+Ag&h3Nh7DcwsPA+tL1GJklK=bqw0
zRord&=Qc^<S;+cjG=_K8QNDqg(3d&;<@If!x3&5xlbN_&rb_wuBO^ZhHo?iFZ3JRs
zKzMO#*31QOP;7074y?JrXB|MVmM#<HS5agSIRm8bom9uY;eDk~8b&uA@$=`Bi*K*=
zfO*ii%V>@Kbp-zwAW_^ZJGH5B4D`38O(}jco+ait$ohn|Mp+JXiY|9VbWQy-@%?2@
zX&MVBFoQ65-QQh0MBQ{^Q~zKzR->NPO_-g~LX$Vi<p3`(4d&%pzlJsG!g!eA3$%w-
z(^98=TxkCN7ul5!oIIe^Ec{i3k<98bpt=?_e!$m35iqIW;Hubnlwk<Su&C3>1#?{P
z;52)KbQl`XUagx>kJHfXXhBYAf|NvSBQ{<&k3S-jya~iuS!P<=N|U*(Eq|U|#2DSZ
z^e9I<%6~HV-xuvgE#;8EFWTsz8ND1h!?hcV5Gmi9X>!S}<$qZc%54yLq3aJp=64nH
zLx|ZlL0Q0i=Q8};*a*2Zz!3)ZfvsUaXY^aUP!dC1!xvC7XHp{CuG`lXeJ#rQe5t3z
z`u@RPCX&hCr$Jrw2?9N3(`SyG;Mi;!4r_jSXz-!6xa(o_&9<2Q1^*uy7$#kkeElrW
z^_H2l%`YABW0Zfbw4MdW0v4V%IEf5oa{OQ4+o3-ndnWHPe5MN!Y24~4kh(Hv{jekN
zy6s3#nCvFAZQRtzbbeJ6O7eB$-&Y^#pvAhEYSIZ2nOytxr!J9R#0jnSfk?fq{f7^h
zX|9$PCqdR*#dC=M0r5qFvXT7ig8uy9cRp)s@_9ZoL{yi`Ivzok9FhJ@xkR32@D21Q
z$cu1N9souBw!c<?&-{vNEPbpReAJ~JYJRIa4!(HdcPkZNnB-0W+DApZwOcoD6Vi$C
z0ukS>NHe*Phq1qf;SC;A+zK0017a%pqng(?Z-;Zr%e#^W^&PoVAGby2-4Owf&C>nF
zZl3ADZ$shz6q*R{&-;$^|M$02`J51lq*Xt0WIjARhK|P%II+5L%A}j(J#U2NDy?L)
z;510SwZ$Hu*+AH@VN}p4K}b#Uuq?3Fh!(00ePO~w6C-`UyFlD?c&S0{&SPZ}qxn%u
zrG#+*hmS}09Kz{1gosNR!RulX6!r!oF7kWNi3iS^8HE%0CgC4(DPDF`DSpO;=GGeq
z*L|NGo<Ts1IQO#Iw{cqFs3ijj(J|_-2M8f9w>EY+KqWg+ePPO^oJ7zJ)z2;cE~4#&
z3l61x#kxA&F~{lG2VpZDv0w|w&`mO-CPuB_I>*?&fzJK9Mg2`e&`~!NVtawc%c9Ey
zL|Vw1^(491N8(Qy_+VU!(y1VIv0%?&<A_1?fssYGex3cz(1<#Vl;r}2eX(tWkvV&&
z@z}w(q8>h%s(^s3tXyD8wuk<Y7;IFPbe!$eNbnE(dy{m&VE0*NUY~gx&p@>}WN)zI
zQ=YIfMZshq_z1Bd;4wq^e&-|tx55NVV1^!=iPaKX>Jz7!jg2~Atx<m`vP6e!?t5j}
zlUEc38b+w;*%kw+KZ(UT!>A0NnH1F4cO^tQvGCc)JRDVBRii~N>R->P62*OSSl|D?
zROT$0^U1n&CV=Q=83Po=)2m3ekb{NAF@VCF;5om&x-K0P*VKEls`rw`t&h{K*_uAo
z3nO&~O)lP^%r`>@713>KiaW7EwS2TtPSS0VLy1f;-s7I?Emw159sc0{PNN;qiQgW2
z<cZjM_Gtt?-f0-5M7XZC3uDt5%gFH&nLt$9D}7q}!2W&h)6k@THS<W@OSguI^Y(N)
z|9-wvbm`1`+&FRn)j%LcjKW2a(s92LPA#aK?5DZok=2?lpL&)GWeJ<)<+nv;7EhpY
zbic^IrkD)5t6o<hgkJu9pLZ=ldf`QlG&&l$?Z^eTEP$sv2Hb)(>0<tbI*KW#E%wNK
z)hJ>HN@N;FDq1P-97zp7&*|NOY@@0ei!C&j02;SK1`FT(Hn_m3_Nl`$^f?bWpjZ_E
z>=eJ{z+rkmJ)!~0DC7Cq?cf#7j)>B48Zzun%_?f_c9_6L<@BYau(u`EItM8{4t~zl
z!#4xoo5{Q=`0xX*SL9<U2<$FJQT*`oe&a&D@E8k>?<<@T=ERudwy(X}E1A44_Tvj2
z$;AA?ttL*lslhG85~{N=E%@O(m%hMia<y!3w2!`5k*L250_9C<NU`eBvyU$G&Bn)o
z%eQH`?J};1Qr<i<^lgh`6?7PEB$}`&zfszr4rUJJQCQT1?J2$62-f$Yd9wMvXIafo
z4=6VP)C7wm$%BHVh4AJB!Bi|uMSF|?(4Vtjwhjkp70L5V5D5EF@)v-URS?)@B*7m|
zN8F@OOzg9hpK=B0574)W$TqnCDeyHU3D_Z}<N)M(S5A8*{w2U)4%1yccS5d}8=RN4
zVHY0ZU-l}>Hx!MtRa@t=&-q}x;~p(<sr7$7WQi^FkVAKdcF*w}ncQN+(gB^MP!Zdg
zRk<YmjQAkJc$5Oz%OF}%!vc1A8gmRuKYAX0M}vE#JSPegB^`pU0+~cwU*Ek7QA#$T
zeX*=52UgX&01}Vzgf}tOdOK3fm;0ji7XJr53xbHE<&k{`cR-0}uGY!B3S$!M$n#nF
zLLTa{`2WKKWEE(g(Xeyo*|*UA8>2MELi%qQ@m%0Rj(FWHW;B^qZ{sReVGBv5KKaW>
zN|Dr%=NmA~hKby8urn(xD0o+EtO_OMc79R@cX2By6pq{61bpx-&X_T`G%RiHxcw+g
zAcuEAXM!J*B#DmWBqE8IL3s+ntcquYt+Z5jX8khuM4wBP3$&&I^O{5C*aRYd6(`qR
z-)_=)Jqn~4=ud}$6s!lm6(wK9gh!}13+s;1g37OMx)w;Q0pZ7qy743Rr2+`7R3p#7
zRUspT22?t~V6c0MGNY-56jQFzSqVh(NY_jyp#5<DYBAkEMmc6o$wupR1THZfw~-8G
z>dmGqTqKD%yHS!pp>0|E-1v5vsWc6TIs%2pQX!!eli*qgI0QN=mQXjikn}^>j>qDf
zyqSdfWU6uc#E!^<&Za0S8jG`a=X<<NwB|^BvPyHhh`;W^x@cf~1%a7=Git%t)tgw)
z^|><pJn_W&;#)kL`CxEWHi(cKTuXO4xGo^+cbL*8I+LyPr$Fqpyq?$bU*ws3ts4>{
zSv-1PAHo^-K4&@UKi#hgYn+unc<>aVwF?7jsUM;7NjhepZE!e6PM|pov&j>5^IDbr
z?Cp9N=_f%2*OT&G<-4`7-9c!1A@Bqa>~B)olEQ-e`$CtbhJ@E{w*URpjWWK1aA|3s
zu|rsj<$nad7Pm%zsyiBpSB+y{J%y26zm-mVNouwcGEq2%oeQmRgaoB2HMzz#`X1`w
zCmJKTuUvwErYDpR##Yp8Q}hh?b%Yd;R$+APKXQ^Vm~kp6#~?wtBb|IFHupyT^636J
z67ge4TF-cNMfTC98zzPf+C??`z7w=C!jI<P%qZcRncZZzou%Goer0|+SqYl9=8#=Y
zxOi8FFRut0Di~6fb{JGkN(<RSASuBK)~PeT=a%Cqst>QVEYhcF<u9E70&g>J0KZOg
zD+m;h!&VOpL~5}R#q{`V)?)^@z}O8mZ9Ko3IwnUa^5(2IctJPu;eSmTp+K~j3>w#N
zA_V)&XKjC&@-aMUG4~mKzZq68y0bkdSV{%3-~c-1;4uVP6Mh>56gql^_(GA?0_e;I
z=UxXyAcKz&1OI=EA}hI_YtR<*BdX_ndvsxF+Vy0^q=qa_UblyZ?>`L?eb4xhIybK#
z^j2Sj4zyk2%^>vg#^|X>&~_AM6-gHNUMDsD^?{f{9a2-wEJiKxo*oto2-Z5~3OF%x
zAe)0>PyP1d%d^Al!^Pie@p4U=U!egBp5Ug<zk8nO^eEp|g+6ZQGeI^|+rc7MekXTC
zX@t7=|F~P6L0oQUA%n`#-4bjtBsm&q>c0waz-DWt*xl#?l;t?R4?({hlcy9!=<sNn
zR=D_ERsZZ;$uFyV=!T|qxgKoR!T8LKo<XE14F)_EPehL11zF$dSGVX*(jvK%`e7Ka
z=p{Eg=&0H3eT(O^CLIpOK3Hh&0%R}qBXWGCGB%__c#8e3?J}WrszD@qL?(X6Vr*O`
zlRUT0HL=J2`Q<wdp>iOTtJ{#sli`#^B&eRqc<U)9naJj`%metm9?5}Ob>c4p`bgR=
zJ2Vv6lZ^%Twy^5MHfStAzTwG{Bstc(kxLt;QA)kn?nAi}{(#UqeH3s(Wg&}A!Pf(z
zxIz%eUZkE}7!WBgyu`VpT1i;tLE*Ezc~};Bsmp?~*+r>UQsy7Twzv}YHu^o!!nuv}
zRIN%{&Z#LT7L%Rsd}AxxSrU)cTj^{9h;KG_a^M<5<E*W_xh!%jwBqac>5NBl(C5q)
z@hMRbIA*02=L{d7vd^_vZtu?qL`e>C*%e}QeXuVF?i)2@Ub$b766cA(lId}VCz>2D
zrHucZvZ~K^tkrPg-;o#P2HV&!VMiB`7-~cXJy5$PYdczid_{){dz9eV`;y?ZnJ5Ot
zP^WjQ2OfO!OjOK^{VlEsO0SYE<Z;igQ!ZP)VMOE4EE26h-5X~zm=8Q^39RL!g*aLB
z!F1cb^L@)j9O2LQpAGuA7|PqdP|id;9;%aWB-f&#(Zerm*B_7^(x6TsA!fblt~VJa
zG!RZ5UDR7>q=hF37@K|aOeLi%p|#8@KCVX7h=cL+-tkcN`LuA711@v{TsNi`Co?Yb
zj?8R<j^&pr=tfSyqrceJW@v}78)>o7Luw7NmA42&xURb8jDfGA$CRW4f~TYsoqfu_
zylBd1M?>KW0X2@s=Y->CZYnIh6$o05A9b~jNK~KWiD8VoH!gscMw4M#Jea{(y#$TU
z?!u2=M>Q@?x6<YPUw$Lrt8XcIAANKCTg<wb5!EzkyF*=|f6A56aB1!{uNBmVK|6gm
zk3w0+S8_@3qaaZ`2B1#y@CkzTI1b^Xbqh9vh@TSUl-lLIg&)2LEB#Q1>;kC?xZu=A
zuOo15&;(WffgC?n!YJnq4n1Ai9q{=*8sDxm!pt~>jLvYqeOzs=!5Z!M*AIz3V}BrP
ziGp%S2jVc`X|FB<Wg|_6d>CB%gss~cY$&6*yu}QG^=M1p`+QK+ps(ERL;v%d&Y$m5
ztXjx{nA1M`fL9wz3eq9^=x`|O2xm0)lybAYB0SZg$C1`;{*!+k^O02J3Ia=N>90iv
z^G^C3g_SEG(E;o%Nz2-4N*P=Iy&jJx5kt)L4y@7v%kny}3NrIZ>S{AH#BgK<+)Cg8
z*B5a{Rs0hZzOeq<iy-+fPNL}77C$to{u)FgwRqP~2$MK=__a<m+HxW>BsKT0y}#f_
zK2N2=%;wLzzIfzBp3K2PcGLWq7{it_Tx(|G!W=0jf3SlFEu=s;NlT|gQ^UxVkiw4@
zkh8jXRT1sy`9TkyD^(=Asj&Z@S9tHI9$*)Q|KzFvqvD=rmrcx6QwPJ!GBV|}G@ax>
zQz4!%jW3u9-4t7J6H_Jddca5cAJd`7;utbK2Uhw5oOVczyf}<aNDe$Dp2*y|c6D(d
z>$9&>O6p3sA&4}*HWm`dIeS+kS<G?cS4hoA5`g=4>Zus=%eDKKsscr{dwf<b)%<%y
zMN)m2T1mbjrC0poofrw92vayL!@6w%5XOo0B;lnFvUhH9I+q)UzRr*Xmu<NwghmYc
z=eE9C&~-9<K`&E=!aNdr*Lg&9wu2#%u!<;Pf>V$w?d1qAh&gR`sq!=#bV}ono)z3U
z91BR64Gq^+UE4otv>pu?puoR0Ewyg5d9zcT_IYuH<<sRh(>2Gfuf!$T3fOIRk&Ha*
z_2>w482a>(7&DeO+<@gPe@*_gNlmpN=H&C_jMEQ4)iBMcXyBss<%B#jcIJWfU4;Yh
z+i~nimIza=NZ4mbrSoOVufJpC)2FYq?9T+|oDL}}JJlS@Wm5Q*Z*feXnW3pIMdA8i
z2+|)ICF|U%^3!y3%)9w}&|*b?b=&ZZ?i8m07F9q8r{Pk%?fpz6#Ajkm6M2`LvFpi>
zg7|FcjBBK8(D9<efes%ESu)tkloR+Zvn@hOrA^gfe9@2=@|4rvc=Ag0vZqFK0X>dz
z2sx<0xzBogg|4p6)z3u<IDJEU^P*f<SF(OC=C^;5Z#M|c>r;25z5Gf(K30o|g&4p5
zkSF4o`p5KT7vA`bUr}}PQhvtZ(Z9*j$UT2cBAUJKVicH8=^D$Fl&n75O!r`0U-yag
z2LhbjDwjf^7J3nBb+VwexpHK6XNQ*#OjP37VLjmLUro}Ed&{v~%;O=h_4wO~t+S4K
z9o@W#YrYz|JE%twCO;{rG%%bEM9RzNslmP}a*-}?BG*NiY!35#X2K=5FC{EG7p32M
zE11RdBJp)g@8FN=b9>?k>)=<==`25p-Cs#SVp36UCTCT71jwN*=IqUQ@}G&;$;fFt
zM6-|SiO0JMUqoma$FVoKr6dh6x#1`llv)_`30{lrbeCYhoTz>Iiks}OIK@=EzfEFK
zA1@rMg1KAj6cx__p#LWWrc0f_;|Pk>#^ZLTys^uO{LADn)Oo)d2Y+>|+^c*j3`8SZ
zwRL01(Bnbe8=4sS88u5X5DK5YZZwD?f!`MH@kP%}5%x22cPIEhEMn5EvOWh|KI`|l
zP4$}R(-3=Do7A*#+SLgV-VXnxce5yk0=$<`JW7>)lxxU>y8(3i?7ztD*cg4QO<TB}
z@W)8di8+Vge1j*P90z(RPqpV@a-JBf!FCyd1x=eL=eQN$F#7-PUh0{fmUdvyoiHtr
z{>=sfdC7l*4mUvqGu-(AvBk#>2#tq7bZvhek54<{dY{I^)hQ;D|4GVXR1bv3UNz(8
ze(*8rE;CU9>z&r=TPr`>EPqrz<uEV;c_*mFf|~MfQq;str3%=IY%!_}=pS(ebG_{H
zb>DGs%N|S;eX2rgdCP;6v5(yJ!&wgm?5%kSZyFZ30EnjrP9V9LHTS&!f$TLV4#@Qk
zi5#`Z&m~_u(|wUb3n~ogp(vm_o!JMZYx=yA;mb!e*4%;L=6s;<LJ!UZ@oc_^!WhB%
zG^3Ov-x^MLEaDBb-kBh<)qJ{_cgH$YBwTgd%R4)cw(zV7OPd>y5qFPh5I_a@3-;4X
z3b#H91*jzlheG)l4d?Ck8N!2S{)Mu~`C}518`L!_*NKP$daeAw#+0p3C<z_8zNJpt
zodalmF2gXt?gvS$Uql#nBpONEdSU%HJ3Ki=-X&_p993V{?1za@3GGfI0a4nJ=lKlH
z(9k}FMi;}4TSqTxdR2YiatB<Y_S$C+Irevjv$zuvv~q(_Zch*;T;EF05NnWO-F<uM
z`EKmcN(k2+zc!v}UcOTOh~edgy{UF0t>zO1-i|YaMeqCzIGdz&AT<N9GxRHej!RN;
zsvIJ6Z4q;GV3oTbLu%;T*m|LUs!afUo#1WqI)T-Uq?aHW?C_TsS`kZA>SN1aj*`wf
z_f^Hn^Kl1qZQol#B`&88^9D}iB%|@Vz0LC8=VoIH{M-v7d$NBEk!gUCqkt4w<_cIq
z+VhykpVu6hYfFK*_`AK})L|Om<}>ouH$58Fz@FjB+*dJ5_%GGl_H@0)Kovb8bS%Ym
zaiR?*8}o?VzjoUYPspj2ERU7d_u4hi?#+afU+s83Wj{bK@qOTzAV9%dSOH%|VDu*V
zNocP;xM!RlpiRp7d>9#>5}y3WUM?ZvThJB~Q@>HK16Q1Y2~oj>kR&5>9?GecDdhN?
ziw|1YUN>)gIIF*5yv2Mw`wFe-Q2}NAHE$T$2Z+7z|LZ`qO*u%RjKbYOq*YORvjCRv
z?R8($jR)J!E}y2m-{jB5{h(;U06g)4KL%I<L@UcX(d|(pe%47d&k5F<Ta>hLSYm`r
ztYh%-ppam(X~uUNl3r@P!$7#i9tyi-VtZK!(j`-PDCD3k8FHD9N%@5DCuqHR_wM}$
zxy%d-9CzLf5rUQI7;;$E2YnQHS>>0$Uul>usrIdHLnYK6+#<%b<6JAV0Yu1P*V}}L
z4KhI1UL=xCJDg(c;<8Mbm*1mcrbm_(9-&FeM6JF`d`ZaY$r`L6BLjHUfFA~Y)%w(*
z&r(Z<E&m<ad!uuT9I@wdm9E73HGzTxb>o;w@au@|1!1=dBfEQn!|hT2yifuzg$8BG
zy7Gf)?g9uzmV{~Q*v1gxe$%&51vvq#WE%Tpt7e7OO&+(@6zQ?LT3w__G1>yo#&dUg
z<j5gSL_&1HLYgK>68fBw-JnVx-}hn2>w9z9P0gRyTbrPq_>`^lZ`n3u?edjlxuHpQ
z%exjEOcYpmC(AQex_@L|A*9~qwfPOFU*Lf3NdIf?eUM>AueHwr_TAHqfS({5Te5GL
zp!O`wWZ6QkF~lhG`yHZQ)>8>lHg6$&XO7``Mok#*myEf!>3|^HIVm?K%UL&@4(oAQ
zdA7`x<jHm{1|pRzxXAEYq2v?S<-mO6?)l0L0a@vNJ4L}2z=*<(Ey}7_NN|1#Ane-L
z-A2+Pe0~2Z4fA9?!TnX9@*pSv@rSiPY>6yjo^~zw3oH0&rW|;Ekl=3Ku}F+@x@B6)
z_x|b}3;_#tfHO{gOtJt6g{TJR)HF6O|BUCCi)kSdzP9ke%<Oy6L<iW1UUxqkYgX%e
zd2?6K?3+9SEAt<@J_>AZ5k{x&1G^SGpT=f@J0(VjN~&Tr;LXjR%@LB`{1Eq^XDm-1
z+;LOgi-9-YV=WPQFEAGifEedyo3iIyHQRULU$PFM&{{$zBO3Q^Aoyi|baV}<uPbw3
z+!M0^H~_!r;W&kF_k9uebV&N-h#d#AjW86RM-DW7U{qMoRex^8wZ~JMeiwyXxak;z
zEG!G!k@59FKpc8RK_KGU+V_AkYRGhJv~s|%DXx?nI)8F*pI}8b=|Y*2e8#xWb8ciz
zpjQYRQ{J=dbCM`}DCJ|xm`8#_uu_l#zTixm6KynfIQ>*oxDNAHEHZbL+>$h`+}Sqz
zlB3)Y<Ci<j9IRvoh*J(9Bjze-qLVoifHy&eS?B9o%wG5p(n7V$qZi@=6%S2FNrL}L
z+glYrk852BchV<Spxea!jXs8-rYLM}77l<;`*UQyL>@CtJ$#tha_<60@bJa?^Vd3L
z!w$?YfR=k|syWQStv5%C_j^_(Q7q(e4PL(Jt2@X00hOY7aoWyt($<F;4-*@J<4&X+
zeCsJZ7#mNlf53>nJv4``7t7;?ho&5k(~woy%Zx;7%)i?IE*wVqz0(MzEyR)dJrw0f
zoodaaCpQ{}lhqX%IiSfbI0h6xos~}s(fy%C*q1X_tB${*1U{c2Z!z3me#ZiWjVDX$
zNq9K>#z0@MbdA_&g3VY{pb-{+*x>Jy4XRw1{en`K38u}SA%m<apsDW3m;7;E1W!(F
z9ZxOAs=MopGzf2Vt*%q$h<V{pOStnE6gJ(<8=pPY1xK5zh>0ohs1Hf|0!vf{siWpr
zgr|hj`tBSvB8DIbHJC42awvpL*o^_<+W(lQC9S6tS@IHEdIU#VYbo{5J{~yvY{pWM
zDkUbA_xN=BCz6U0geU&Y)w~}MRXh-{WHSsA{L~0CV(bwd+w$;H!EF_tr?F2m?;js_
zv*;JGC9&y!4FZ(kI8c^BHMlhL)Wb>hjH%yj2NncoN<=s@HYNDibAIjkMW?_NQSALA
z%nB3eAuf2J1W1B!1%Fio3bvsLH`K($LGao!!lH`9i3$AhOHCgjc0JU1=;7_rLA*jz
zkgN<hzX(P{c38Lc#qJqrEaewvT+AdQezZXie0psfsINQQy5*A&=&j{Yhbl?K*?%oZ
zuYqqP@#OetM=vuYR4oUNxIV0Xy8YSCG(U3$y~jlx?BEQGhri)DulHM_l=!Yy<9-T=
z!jPQRi71W%xc_l?islC2)yFw}Sixmn3Y<Pk>`OC_+fMk2c}%_5E}H;1dxAAT;5qU7
zBPSkihM_;o7%dChR%*>tQYo^LraU2K{IA!fOWJCq7rm^%h5eWGE7Rpn!Oy?d<h;KE
zM$|34k1=zyd$Z<`Cb^-_sm3}oT2E^-tZNaf&nkx+hZ^{W0hx5_KFD7*QO_^XxoX~I
zs%`{M9f6A7`PU3jEGD<#OrZA(U7v^#7cRTU<OTu&(d#(|g9vrjzi!&zd;-8{L9EFe
zVYxh##tcDQZ5IQJ3hD@(Z<C1_mkV~(oY%Q{_~nR^U?|YR3C<)Qio&bjn+RnPy`U{`
z9R5ccyn5Yhwx_@iZt_9>eGk8N|I*f4Q*@it>2S@eo@QQCv?MDiQUbhyV`g_4lfoWD
zxhtgu*{<`v(l`GuiwI?cLO~HRM?=D<;;~JVqcp|r*99y9%38Om;I#K}t-^`OqkKUr
zxX>S16OP{>*M$~YOLqD4f#^XZA0Q^yKto?g6fLO>@=uBA0@M3y@FtRt%Mi9^6V09&
zemghTtloVOihzp%;U1OETF*iX{!~aelqE+b=WpRaKB9pi7%=sEc3V>TBaP0Cz{lZK
z-}l_Mv5VX&{L|VM&`u9M^S9!(V4V0NbJ;(Ogep`N>PWwujyxfaP^yq3!+T%Y8>hMM
zg6cf*_NI83qYh^z&V3;@Nw;3kbOVICfa<Gm&c^ZKOqQQWM~*QlOqm13PG&1S4goNO
zV`oQM%I{RD)19x0^e}km>d%LvdGe0e(cWq=DJ7Oa{)+aT#KdAZmky%Z198oV><USC
z6y83q@-7F=yB*O$KL$un90V;-hBk(%cfRzMu~(p;XjNpg*+))@?0-T-CbNNRFa%jq
z66w0MT=UZXUVHS!wO4OGpK0hsCIvgV(?V?6`bD`Y&t^Jrt2f=f9@+O259n@p2qdyE
z5if5(?IhI6U2g2``0#M-iT^pBG8%GPhUZ)8iKDl$YFkniv?+lc{IkigpD<0aVv9b%
zge0=&9XkBk5+!=B1*`n6xyC(98nH?^y4-Mf_0=9bU5=vt)n)G%W?;+6Z+@K;0i2X7
z<jC-hYrDsz0&^EgRF|?7a?`Ku4Q0pWv+Io#bSPbphSM+ZWt=l5w%qrg)}}Z6`1DSd
z$4ew)C@r}LeJ(oWo+X}H?`tsV3=~OO(xu&(IioMX+~=|){fW8%TK-%oXXd&uY8n#%
zyraZ7F+}6MzrROtYF2o~8X(y0mhjQRd;FZ#{PU)krpik3$_L0EWWll$n8BV6{}5H%
z-&@Y7@{+~+JS~#8GrGPhq&xqk_J?xd<o0&?=4XnUiv`g|hCaLDy6@VNC6c~hbZ%ch
zn}7mpICd_$l7*?QvT?|7rv1<)n&kFEmT2;irjMI9!>F#HR*rtLtczG$u&3pUsgh5a
zU%P@B+KhC5gV%)|{NQpS6AS_J3;8p~*T1DR3^&<bKn#oBykvo$6rI?w2TQUMjF1Ru
zBh!+dotq(-C(8Rq3Rel4otTXe+13<lz237gE-^M1!3Wy-d-cWNRWN_NRmu2`!MhN)
zegNw~3fFOdbfQqYK7(kC-2>Zigr8RD&>n@fkCmQls%T8he*z`Gv|0eiuD1vaHb_${
z?WPKokNy<#rO}7?LvX&vpz@+JZ!H;<w5}-ZPV*%}iqR9cQi~O3UNs(!Y!6!apESEr
zu$>$o&UorT4gqi-n&|1&TDJ7(`R+9#?h~zloTbkpaEcTGf?)W1bwq$m-|nzy@0y|)
z83HLr9dOb-rhYX4Wl{CCSnp_+nFB9s;0OaYF`uR5DcjmV2u~%w7ceWF>2}E(p#VkT
zk1bu3q18?b@qnh`8oPK9kmotssyJ*bKKUzIL^vLFzIXBt9Q~bzt7a3)X<!*h&!nY$
z^yeH`znric{*;UcO0)fMIWj`UURQ1XF*bR3e&Usn#_YB-cte{pd}KFX9B|@u%zy;t
zO0k@v1^TX{n}b~Hxc}RxKruP#;{xXJOMt8aEwpH3dTC*AEdVYqTbECZvNS(hnIP=$
za*YNLa|7)8a&73of9AeCLP*X<<}LqsrI#H#gY`;diDZ|N6F~!CWD&UVNBJ3J7Gc!j
z4h#azq>bK>BCIPO`LjS=qlp1)^hbNU1Y0h?liwTd8w37pe2qu`w|4&$w!ONKx2pVq
zG~kS3z2gm`&@T$QaE!u_bSyt4XiokeNZ!AH_NQ>;wW15)$wbQ)COd)K;Km&~_}Qi7
zX$RMev}Aq!W*0917S%ZRX1G`$u043k_oz=%c)sB%LxFJom+L6&sgkrJ8D2Bg5X!%V
z<hh(9hiJ_6<ThztKe1`{Yhi?<OiMyX5wCaya)5PZe%;Ca6|~B@#+VfR0^;|ZRfR5T
zA=lCmC)d>oEjCHstBgm#p02-#E4uHpjI;(mBCH&85x6yfG3gx!1I=fbOkCi|h$tHM
zjft9<Q97Wzf$p{7AOfQnto#f^46&29_B$QtYD^6-NG2k%2(>?6gDjqK`X1|4a8#mw
zbg;P@(0IdW3AI2Dt1t0QK?S26xxf<nlw$>rhK__=s4+GqZT&_pa^fKuTou=`m(S8=
zQb@MmYrmB@iU8N-z%xS^16Fy4-d!{KxtieJ+w)r{wfQ2c2t(!@F*3Zh{ac_*hL@N~
zT|6r$^o<S*VNPrLP&3zsi8<fcKVJBKl)ts*ZvYt{BSt(yy34wJoOt!~PJ1I27hx?F
z!J!zOWu|oX9xXK7i$(ZSU}Fa9TO_|fHPB~3x{{x~a5)_GC%PvqdFiYNDd)tRw@1Ys
z=%k)u#UBPdbz=V0a5war79z@sed9a9w~YQb!{-R3<$iL@048?u_I=Hwn0JCto@(b_
z0rTl^^PSSzr)R?4_u-<NLLp_)$;d_BKJ0kF(Z!Cmj7PMW1%UsL54c}4C`-8^A-vR8
z=xp=4&^Jct0wJSLA?@n3PPw8cnserQ#%RAes9!e}P$^gEEqXbMwXi^nPoMJcwTxFw
z{$daz4fl7xUKivrMhnf0!GD94aQwilh|jKqH_5OPM-25#S6p847qd*>vH#S~&>qw1
zl~D7T629V(SbObC2{TKGLYK7$;8f~Li42$;t`Vl;HDHw@8K~t3eScJmEqQ9gvnkr>
z>!18M9-i;^J1en*U<|@09{>4CAXNK`Y`TqC;htdNVyF1aSv$yIF9}{z%iPqZyQqmW
z4<|}Ox+2+G-eq)wn=;5DY7>#R@Kj>^KP-UdQ0ZA>OR5(+p(|DaB6sv7&N2;E2L$X}
zv?*@-OkA}puyYl_6n<`y(W;l1l|5{0o*Ea;cvdM<d#{1tf6ter<ap3krne{L<nJoi
z6~(%T+(X3*j$aj>%|f0{DH1mCZft8zs)q!@YhD4Dq4aP)6V5xNLH!COG35jtkGHwo
zbTqZFw?%77Ew&sE8NQ{U<+@FSGOgo1)V3&^b9{GEZ%GK&q@(#GPJCqvoDYZVrxn4m
znyD!>yyR%F{V*EOOZpz5PGvs~t5k_IVeQS4JaqQyc`VZ^f$q=pj!sAMt6+r}FCuPt
zK$-3{wV){s%yNyWU@^eNPJNn&A0B-#ity&(+>oOG%x=cXa>MFZF2b^vKcV<1Bf(8K
zsHM`y;%Iw^E%_-WM&lQQ-q+_q^95;zQ@}9oBOm6x2SB59>-v`~i&bjR=$U>iV>3Qj
zHHz-Zb&AWoPFqueY>+68eCc)EzkQw9(y0vPOjBT{40yjBgmQg1x0B{flZ;J8<O{ZI
zf_m(}`M@-fpRwXr^>gc^2E5!^zozHZ@k282Z7J}c+IN>Or_4W+47R>cjsOeT3I{U0
z{N1GS^vtK-g%r`8zol{l=1m^HSx<WAZ#t>cB#&YFes%la#C=17gJvUi!uUDk3CcA^
zE%Jg}%3$WRotlw4=svIZ6jGUPV(TGvqP|Gel7SlG#SMS1BitqO`(`*ptVcq9=n@k;
zO}ykDvIkLc{j?`r(^^zARPm-ghxV%yxc;&KCB@g=ZK#>xe=o$Bp5;x{2AL`G32A|n
zv^^PKYR!Puhjc^$+MX;vveuBT-tI%=w^PNC#F*z<FVjCHKT#I*Lm)tbQiUDd8b=I$
zzJAtpd1#}t=eZiX*9=IEC^a>|t}`%YZJZ`M5iD^lqW9+1>S9BRQ623`c$T15B1T&`
z28?G#vF^@VfrR<rlkNW3Bhw0R4jpj>gt4^lpcp|O=>VLNa1&ZUN*Pnx)+5SSg!oi{
zSux+{68RqBUe_DrqVpS7<tCjmE)GW*p#C++GeF<WAd|HKBQPg2{8QKF<jx9<+1hRQ
z+sg)vQwLN}0FepLniS4h1ZBy2xs)QB=lrEf4iOPLlf{0b%T?By-Qs(V)kdbZrVDG&
z`ndsK50?TrJy*^EB^<%Q5S0Y2sgG&91kH1|Sn$i-m@NRwDoLBbi)t@JHCO5w$rDPh
zCfLyU!@N>2kr^^RxwuTOWXPB0*{wjsiX!n1a_FRmd|cstf_y6Mm=SGRwfp(p!v%Xi
z{wG60uiTpd44My3zj}s;%k<FD!z@N4E#HFmumRt2A1gIv^u=A%*eA03Jp`p68$6^e
zOFJ(3xMHn|h<mVn*|mD;U{2H0X1F~18zj%kCxfi9&fPwRC%Jdiy9g?m^B6ac;X1*f
zlVssRAkakVz*!lu#YPx{StbK2!E5&`&S<$e*fgk()zxk8_>Jt66|sPKJ#VeN3W<Pk
zkgq8QFa*9rG|z_E*OH^!2!3G~OE~K~ao(#8pRcUD#I3$6{yIdCmt(v&(69vi{dyVt
zH<5@W@*Onx)$0P`*FW>E=wJ23nBT93X$Z|)RQgjN5}zY(mFr#35dXObmq4B)UkD5Y
zogrx;j}fy@rTUDg((~&oVl8JPBv&S;N22yUBPgaH{{1GLrPWK7P)R1wkw*$Y<QNKx
zG28SH%WESP+HIw&*TpHWc5m>fAFyk}nMB-XaTL+{Pv;kgoae?F180f0o3Oz4nhuD|
z#24_(jqH&jN+t;LKlcS_K3$eRO>ODPw8vEfwp&k|83PXnC9`uk&Q~>6-4_IItf5Sx
zlo<+#bA~`yf_&jPD~8We!lrjxAB-%It4B4O@l>h|-M;<9x|BWk@QXR#6>&Fe4;fuE
z_QO}~pGeLVA(F_*5ku7mnj`a66HA6HPxw$<-4hL#m$dT7L{QKTR==|i-iY>C|Jr-E
z=)KQrMNyunIp)Smt_VjWvt=92fdqwHa@IGi1{>{V=bF8IW(C3^Ii3#=lj&mk)kozN
zi|>PcZBJKg36Yn4g=Ht-0E$2;6AxA0HShF4aZ^UL#`x`whW#B6e7Mp4$Ai`EA$z~%
zDPJsz>X6m_CNr|bK|w4ZVy)Hn_`F3v$&m#uwBc)E#$wsH+7Y_>lUeOjm*KPN^e>K_
zneT4}|D*MJ<d4Kb*TiqX_HfYu@Zz8ms6Rh^gPikwZJeFnjYo9NkN$KGKVQ#%`q9r8
zmYsjK6G<)$ls7fe_;KZT7@D}%WuWK07fGwgV|Sj-jNTiFAtuPP$cY#?>%`&pRL9uI
z;k4J<w<o*j>1VceUL!D1<*(~eeEmJq8k`&>WMmfYv6*)wAv()yR(jp%Z!&y3`JVF~
z%&dGX(6>`Dj%~bKa0s4h_?aMD{R9e~Ee+!geh}eaAkub+fh`vP6ZAP{>kaG`s);bX
zs?*t>p!h!PP)IfBRW5o)5vO()tW?mWO9EzRBU;7(UvFRj5A_#4d}j>C*v3w_u`Bz&
zGnSH)q?EEm_L6-|n75@6q7urk>_V~&V+oazeM`3NLPE&Q^X~I~p64HUUa#ku_v<xt
z&$;*9bKmcC&pqd!bN*`Acr^>+0N&6|CR~Fo=8|p9>L`6lSgVq5k^DOfdHWwZ9H0it
zbJiMFa7U%a??32kbw1NK&mlhL%;h6Zt>P`e{MFERGWjYpTG@arayZeS_MLK!++c%z
z2t?t+<jZs(HCZOWz>D_*@LdHClu#*LdnqxLq*b4dh#;~ck$^5DScUA@92jF5&hxY*
zj>q~>mT3R#1ENiOr_LFt{d55<Orp(KDnu$H7TB^59No<?Xq6=2nV?u6LWs*$h<sw#
zdX+LHu)%*04lntu6ct9URY#;IS!8m;@N|_d6r+AQ9d?Vp%rddUUTXR(q=IHPp_sK&
zKDq>vr%$OvoEHLq*YgNS^fhukWB$z2ZUCB(=4wB;5DpJZ_W23N?>3t*f4zA9#>qla
zN<=EFbfa~9P10!AO!>s5?#qHY{({UM8R6?^y`?OFMKWJUAw8u_rERK;0IXzEz_V+U
z=j^$8P?9#_(L;u-JkF70!9wOYAG+F9ZKWGC^?mieSwBtz>3{LzG<MQ}sS;^=PJ(hc
z7YBUwD+*&XDn&0R@IjBs_J}}(z^A&wh7Qbo0QnXEoO@!6m;fDxFBJ-KVt&SkJX30#
z;}OyT90&`W8h(n3pSIq$bh}jFL_8L*b&w+^mfyri>#wqf@|!}vF09DI9hvl_6?m3u
zU47)u(Vwy8)6cJBinU;i4s}BA^Rz<?8H#A<_f>O*r}_~GX@*M4^Z}2LlgbAjj-`Ro
zUN8d3ihE0&*j9c2#5(31+g;Sm#hc}Kw{=%_jz1ix2)<#x5)lq-&<qm4=gE5#VW-n1
ztUkr7Sar+)eENpCJM=aQSZOmpJHc^wI8T$w_rjhF`>V+xcZhJRvhfoVMdUws_BWde
z<gRiKvjCl=<0FFKP1k{Qy4B)n;yRpOhWY-x>^-62HtWiJ?|WIG2=YpiX3FW7@IzLy
z>%1%dEXxt1$chj91dFrB+nwZ;EdH+{M*n*WdC7$POYPh$k2AOXd-F5z1Z^y)e?)xP
zP*PTwVhH$x`gXef^rO1&+<^Pa>(yP~$?vW(pL~>S^Hk>Kx!f~-jAu24E(zx`eUA!0
zNElY?Mn79?%SB~3br$~aqJd@=b7OyPon6*lE>&%^YyKO^zac9UO{_AWLpHEn`?Z$m
zz4JLDJ|K!sdE2Ql0E0BueBUY%hhKP5FT?eXIx#5Z_kgZ^<J~8I#zNe^9E+ONvwf!0
zh``Re{-Q7d<;~JF%%TLPKg)oay)^soXXW;2s6_leir6TA`R}d#)`Ig4AvX>;G`&Qb
zpiIjy`BgbP`y887X#BRZ97FfZkF59Z&iFP?n)_>HhGz6#Y4IB+Hz8+2m`T)$L#XUp
z>i>gl)8!5aF~hji|8N&rAomS)b#lMrVAQTU+X4N(Uw+z+H+1NoFYVFb`)OITxNnXk
z@A<CO?sU8e<#hQUR2v)`f8el#h_Oj3`fMo69@&qWKaH#2NR$Mp3fvION(fCk(WOqg
zx^A|hRt_HuV>PCA#-y7%qHdByM-&M0EGwif9hcEh#l<{FRP+cWg2HdH+5TdOm`dI*
zWNB=L%Pdl*daErr*r%K~wh;Rdhku&xt7oE3`s`M3;H`Ho1qjD#K!qW$Q*a7_4PTj7
z*Vpjd?Z}J>oFeShmyMTv3azOSDBPz;Rv_hJ+jPgDs>{?5g+_#o4_HH}TbG{U0LfhL
z%#Zw#$k`}*Sg5$(08#$M91*D%i2uRjbiz%W7_0l(Vxv8VS?(to)&R{1<Bnv*(pU~s
zeNtVA@;crrp9w19572ub-1kY>cnl1!Q6MZ3a@qkkFH*08@EwXF2OHO87FL`eXW=L7
zr$D$CXqF&W>~1_>I%{1?l&EK`5D~WE`&r0+5iK|n^AM<ccNFHta(olS+!k>z3uB~2
zOINnDLZON)$eA%_w=vIO&W%lOq-)+1cvuQ~s?)!^V2}%5Kg1jvG+>L_`lYt#v#xWq
zLsrtb5}!$K$M;BV<w~D_%CqsCAL9W6P!f5RIbHchQ>oTJJ{^5`5sg18G7mHDBk~||
z@))oiZ4tFR<P<7Yi$nodqAJ3(>9G+~Rxou?xJD}zw%VFPUhVG8>*Q1$TX$Nh<*H$h
zvh&72QsS6FEdgf(Z;NK?LN-VI%%BBnJVdizLH15;*7+1kCvpU3=77O}G{~BPbc)lr
z*r0D->lGeDnK9q23KBse#|!RZ>UCByK>92h#q05PPp6v0f^vggrSLwP8~^&PZ;fkV
zEu7H}iQLl>2feFy>A;|rcR|rptHkAkFFh?}NRi5i>zcYz0HSk_mlz@+)shXSnCz>A
z9u`<hpx7gV?%;p~ar4e%Dmn!ioWvX3I$ONUbjX^!?h{nZc@6OJ4z*H&`&46F?$O*c
zI(6a*7lLzrbbqjnxCP~qCH&&@GF6Z@Cuia^YCZ@JJYtxzhS!Fp-5+ne=Xvd3!8`Un
z;KhN+``P)*u){HIvv+NXim}UNBnM<u?@r?ZM$94O`YyTRq^(-DGb^;^jr}$>J6$FB
ztIYr*%ZvZIpIjxhAnV^*KCosem*!^*Z1tM7vCsgQ22GqRwdxy)GfGk8wO73B)V<HT
zzi*S?2q(^FR+DeWtKFq{KF)fD*h*#vY;0urITSJIQx1vWQl@7g_B0q5s-Tgk>IyW#
zzBT9I2GF193$teq>ka+BG@QD_iNI>{&Q2Rg0t?28<Dnw{jt2j^0G3kBstBAtv$s$n
zTkfju`vJ}{IEiuWNEVA2%GafZTdJ6{--h+Rf*Yb@K7ZqHYA>O}s89t34^t+voBM4{
zC54GoWlT}2^d6U8&xX=0Clc?f_wrcA0IV-@De!P$fZ`B066WY=Vo*(nV-WOfoGeW4
zQ#c8NF4TD;v1e6`fZp^057M^o&7YI|_Amczpd`YBbaCLrybIEy1~I>MLfEk^z&!nW
zp9m;@Fw;G5U$oyA$QtB!an!5Tc>J-;TrEo}^(ti;VC|cO-5y_!%phYVUQzh(N6{K!
zMv437RTGhnG-_0*#DVMwK{{~P3TB|r58r3Y;%;_wT%mk>!mkFI8f)Z}UoTP%f>qQ$
zQ79U}4q#l6s<#Tqf_}Xw8#ru`O9IRf7`^#YU^QR8dTH`g?O;2n@@Va}JsxNVyto9C
z_|I~Fr?KBW@dASYQbTK$umBvJaLz?|?kjgabLDjDbN@wY%c*RHh_n}#T~dBl5Jn%Y
zHaE@;NvCJ9puv`I6YMwp#Wjqw&;z;oaa<{<Z|r`gB>r8mQzqF2?=%dZ$ZXSg6BT(_
z@cYv3(gEL15eu#%OKNx+b2Oe%ECrW0e6)fhj_$0wB2bjX1K+75_Bv_o<D0ikcWTQO
z7|bqriUta_s^?5=z;Zo~35x!ua}G3|5O>Z%8S;E}(h_{#WvP%$IWzqSH$rL*#n=%9
z_w#%U+i|xl91A%Zk*3;jNlP}_Z$EmDrd=f%6rjEup91cKYn5b8gr`=uIHH0engy`5
zXJYMe^S;w5Or;#F!x12<FM3u@E{|X=A3<@T5oS+?{J{i9z57X)``>B5<>>r*K^Lp-
zhNLUoTckz;)}h9KQ@V#Dy1r}wXj-=9IsI_-^_X8MzM%G@920Rc0((zl2phU8#?g*x
zX{70I%<x>dO*8d?(=rF2^FZ{oB+hXoE~B9vCcQemJw=E^&({F8C-zQK`PiPW>qvw5
zc&{*rWSV2_Az-03RDwZ>bE=8=%Q51}-bB|7U?QU`bbo}6>0t)9F=wXG=GFIARKnFD
zysUWotjUsxz&ii+gi6dp>wtzhNHpD)L1o(}`DIIt4V9~**<SGR#^}T_ZLl0Yaz#Ok
z?+sQDP`FebHj}`^<)Z55>)N;w)2N3?h-y^{eqZBG)kj8Y9dVi(>Y3~?_Zo2_`^oE{
zS)_oVFc_-ieh32N2|JdFl`>>TB}8W4ut%+Q1b}gOAAqC;!dO()^3#jgM6_rOxH@h9
z3)rDZnpV9PXjAX#exM>tmuO@0?iHK#2pvipsS27x6<Adfx2~@~q8m?E%2{YE$?(NA
z57m#am!;kSS87k)Jq_y-Uuuqzm#s%%-c7W=OvQOU@Zl0aUCy)^8lU5$!DK}<{c#A1
z9dmh-^>yzN53=8qa;W*W^6{yDS-*+>b}U$<F3ymVDGcc`unN(<#mSZorDC3R4vgpL
z*zlt9D`WbaV^{5NA<*Lmd2noa^@2$5g$^N5jG`?AW0rONdE+jm-=&1#39TM$Oi^=$
zU|C^zo!iIuG8uPItPaEcichr!77UAvE+3KoA)G^PlP&E-)u7Pui73b=t3HS>JK#=v
zm&M~HE}$g~kbvDFju|q(0L)k;1Gy6+M-<*7OEB313uAMadT8~{a}q#fZI)_M=BZ@}
zKtDL887^}87g(P)^5OjKIb;{;xgd}1->DBQ_w+|bHy<ydp?l>oVzqthWbVm&m?&dV
zK$2RHQPn-{!Pg7GPDXNT>IIpZ(soPjmUXDWmzI-2GVgw-_W;|{$u`nOY8X0B37vEo
zVfF^!mnD~ePgr&Gb)dqJW<je8!QPT>&Yu+#3)TAt?)UvjS^b(|-bsv;)EXEE+l683
z9;|RVG$v6et8t$MXFgmdEHftiiWn~_^|&CSCEj$juk(dhlBYzWiDbv{@IEHsi!g>y
z6_RoFr75qHKG$DJay@&lJS$yR`^K}pTW1XAIh#(nyar?6)|%My8oq|rl;QUXnVqC(
z8S-&r)`c`Tc(I@Y>l9M8D;tPW)6!QwIlDJEvPx^+8gl;}(pKj4-w``v8q>wuo{Nq?
zG17nR=GyUt4o*IJKs1Yf<$unUTZ*BFZ5DHm1q!|OCGetdSqhJw`qOrcquvZLH`{#D
z$Z?fHpxFo<CSv^Z705F=W6ceZ$dhp?2L++CKG2E#d_QBy5CyN`vM9bZZ1jGU_+al_
z@<@P!Ujj&E9PJHhd*wvJMI><!?q-AQFo!<K9)H^N<mnnRD~V53B@eBeZMnZ#Q;3GI
zv{Y0+)?VP<n2!msR*K!WeIJ`8s_9CSxm7?(nHnRB^b{)#EP6|kuiI3Mg}mibuoWoc
z^@a<v`6{f@7e*C@_>MMf9#$D)9`1AAne3!JQY^P+Na%>431Db^IWT3GsnwYYun--r
zA2BV5Eee{}#VfaUS2;+Yr-79JWPDW0*r%H`VKbgHO$!`isZ_EZSXFkN;L;Dv2J`Td
zfLS&~C+-cZ>&;oqa3CIWv2=V!{HXx>iO<zkzFSIjC%>!$vm*LT+HJg8&+r*|qR2#j
zIX^8`Y}J8+sM6p=NpaBk8{&Ws;7LY(C%HU++kfgGyOpO9i1a^vEJlgoyy%Yz{8Pa)
zH`&)!uy(C>_r>Q|hn#G(^u_&f=>y$CsmY`R)0|%0xSmi{?J;9{0=5V=J)hESos$Zl
zse1q9^@n$2X~0cbKF~N^{j|26IMKqJmUvv|`Y4m2Hk}@#$;UW17Z^Nsvy;j=BIq;@
z+|iYySJ!{xgu(X+zL2D&giFdNwVzzMq{LT0rGwX@S=(Fj`Aq>|rGPhEo{qH1`N4uu
z>7&$HW#pX7t$^wW3ACwHyN~c-<O5H{Q5_QNj(B|v+Em{^%p}RL=p3crmNML}*<xP)
zt{u|5-H{E1K1zawqt2Ke-w@IdFTy4B9m0M|LYKbdq|m=L6@ZNh?S6=RrB&Z0?@43d
zsQFVHRc^N$IZ-mEZ_=wVkDMe&Nbe?J)ptVFeK!&_*%(oK-_WE+&N?=~JF|H^pie|-
zM*N@713OuE5yB@S55Gat4q;@!IB|e+I^MJQ8g4OqKXXL>K#NP16>8j(iHe<{oeK^4
z!2#bf5HIBThdCyVgR5tcZ=<2|SJ$FkA4V|i>p8F3aOq0K)d0GH^%bo*j`H{AjhEo+
z<bPY%WfD}%Lzj6p)*)9@88ph)(Ri~^cD<p6u}+h6i}m9)o&0uqV(iDJgLwh5K(xor
z^V|DZln4Egqr8nR#l`#1<uY^srp}Lr%mtjbu2fJXyc!U{Xd#bhG9Ec}Q8q<yf3I$x
z+bm+0r-9~*dB2U*E!7uM-_!mX|Cao3%uQ;Pt$FHN`-RlVnrb-Dunt!b|8<`uCgth-
zW>^Wj3x^H`?Lvq<O?i6VQlKk}@wqu+$0y`Z=;)Qr__rDVrw8jes~=%3Eq0oEdpdrn
z@VS!F#D=W5NZk7TT6OX=WhRv*HGPc%!7prkFRNxZw7goZ%(Q+dQ#hCu!SW#4imFVk
zVOrx95)H%?R`xPirKGX!@b)Zw`WClsI;iX8R7eyHs`RlX2Yr;Vf*4@1_?2}0$m%V%
z6TrkW)Av&0R9@!L5(d`d=;H4-YwLqENCZf0ofg(Iq#(L}ekP4!f8XH~bmwgzA-qSH
zoe^b&1GXBnqhY=Mpr7wiU2HgIUz2Gfp-)!pqT5OqPUT16U2$$S7-Q(9lYM;5zlO|J
z!r2L!4)#(@2G98Bhy<Vt0u0~5gG`2LQL5^*WP5DTOL^APt8$agbkozq=*#sUaQ%iq
zVY83ZTy7_3U!Z2LmpeL@-C0QlCU&fCLx8E~Sl(*oG>0T?d1H+_G56JCrgZ!E`Ok`E
zuKWYcan{SFNk0BFy_E~q#~f{n$CBf+g5p(M;ius~2e8#EY!^hd3aXjE?h#72Q)0Mm
z$eH_w7r9(?b{avC6DUDtbp&B;k=?G6cFDcU^r#tO*>#^@jxQEop*D!EuEAo+iYi(q
zxj>~_0?y4L)ePbyPQgakVY`W?QcSj_bl=~l%fLpl(gR<M?C!N$d}}I>2C7Dd-joOn
z-a*X2%H|h9s!yBmk-_jljJ36u5OnL`)d>^kVk=!Ra;6WzBx`<`r&1@`gA7-gkjTsg
zRVFzV;Qf{dE7d=)oq}UHss*M`oC%6+xgo=O`PjUvRIrn@#0+d$vPUmZc`%?*&@5Fr
zfavkhVej5V;TOMTc4*n(Ho>1Pu^UslV>ijii$(&#U5tjM+gWx5F2yeT3V8g3;h8iz
zu18=2&@>f6BP;ZRq_eB+x*pfp#EKKq2-!)H`CvxyKWdqZ0gM|BCwGjj7+d?Mf$pTi
zvs1zbhz#UwcQ)98#vOJ_2Ma0PN^@nZU47Lz*QtbL=Kv#d<rWCy6<N<mrj+miXCiMf
z<d(!3O8w<~d0)ea8&YF}Y&2imFU+IBB7eH*mA7!Q8s?QhqxlVdoJkkJq6f;rn2L{Q
z@Tx}c!%Ypw9YzzjVpEuH#-pg{RciFaWPL$I{8gwwh}?5*Tfr%=YF8)^GhQL)a$8UR
z6^6?i9xV8Ckj}-&nz3W2nl}}DbYO^EFocqm41UU-^mvAovb#<&(nI5HmjOr)V_Zj<
zUwe|d9)PsH2WRny;q<PP>kbLsZu?i$QqO}-iVGkC$g&6T7;|cTJBARlIT-kE#<#-j
zKyHux<|ddMq4*>LwwIDrByLr1+HzgGLT**g{tqU4d3Pq#wTAlfuFFnK1*mdvM`q`o
zK*^G4e&?KHfIBguCns^Zy!-Z3!u7;}%TCs1L_RApypIokctFh!ol=8ic~~UCVugSm
z*H+{N8NN8fhrsXhyCI7g;GN+@>8?B_@z0+gMY7&9{fZbCY>MN^4_1(1hfx`6?hP9N
z#J*AtU4v?SKDC&J3%Tufcl`aYk>Xz?cUxJ5k6W_{P%l1t0_@P>HB>kk`Q51h^{pJ~
zxXofkg4I`j5YCd|M|^#`j|Pc53biA*KFo$`pAkmCtnBnskX_<J;T*;-=XI_k^k~OK
z#6geF80(eRFp!S2LoDAFNtXmcBZzCDV@6-RZZ+{I15NMlNu=ySITCMVN=c3M#6)MK
z#`FWsDDkz3b-CB>vD2Y5s^JE;7ehw9v0;wknedPX(Te{*)qhm4MI8({vV&@|3;nO}
z4EElp{2lIFl1z0?=5E2CUmJv+2du28C$Ld^-=$8KV4>Z?GPH`SdyyVHM<SDDlZ#88
zBqzb1GCLsr$(2YdB8h?l3ry8F=Cmo&lwH{A@;yIH67)fxhT6-UZLQb0RVi8#s7hFr
zeGodkvx%NeBPrz~h%(#}BrBVlRxXnx#NNn7?_2<A<Bbe-ukbs-28c(77RhSZi7`X@
ztB6u6%6E$BL3&n9PfQmtiUF6%1`B9DxvDR%Gu}p7W6#MOsdYbE4?m;NW%VerSM?(?
zk6>f_Bg0xy=Z?F6oZ!USI1@*)N>|J4?|HYsjoPH06!9Kbu@2DHQgn^{I3r;7)<+Sa
zI9R`a$zWYBuWS$t5;@So0`Vm^nDo>33nz(=9tgF0zDOin+J`r6rgs|TYeRamV&br?
z+5(|RkD6-^-2T>*V?tH7T|@oA|C!jZgPjLYs{sz+Bw%F}8EFPZ)8=mI*jEd9yl^37
zn#56Gg|MkNl260JF(Vs>vt?7X$*7z0a>t!ns6wG#e|_0Byk61G01=-uPQQZ*dI4*f
zLegn1`;D|V#TQ4*-L=H{zSC?^vvtgHtZ6(qM-yYPV&L{dqCHN$+MP#XLideLYE-;Z
zY~)729Sxpn)WHPC2t=`GAU&j%4Cg+Hz&exp*dg-qk<2m0m8$}vKyPYq>y3>n%C^o8
zt1=y`;!*26m2EY)mn8=JZ0K##U=a#K!~48Tsgfo5t7)A@pouE>ZDTm&$cFS3G?3OS
zF{HbMk)2Y<ox1I9J~hGU@ZcTh!sf+nNKMZRg_|=WE`AgO&Sc55$6S=S^<&tK4OI3T
zNd7E%VOmJxG}&M-dZo>T5-mfGIbyukEer?~qn#EQVtZ5u_Is#1-2u6Y#kpm4HdEW&
zIrO+*X`68~^3URT|CBSt!wzd}b(sbn`A>cG-r603-sZ-QHf`(C)}VuIxXj_0x&sT|
zSO2`vIN{WaP0se@o-WrQe~^%o9a+~zo`%@f*uY=M1RYuLlwA$)PhEcWS$QhwAZQQ1
zJdY8Vj)#&#H?F#7impu$Kop;q^i>d#y}3V*J$NYSb1R9<epB`g&02mu*FD;w7>Xmt
zxo#2ArN3Lw1Vsd19v;*&uc?>+%2E4##r@6i`R1}Ag1LXQ`Vp+JUes_MGADMfuaS%A
zhd=UnnES2V8)m5cuF^Geb{2^g4nbtEx47;IMR9!i(IKhWAx5C#_E+qVFxg=|Vif!+
z3icbX)mDSE65o9nKJ{Ap%~%lBkM{a>_!9;OU#ZjmJb`eb*rS~B@)+78A1)C>q*k%&
zW@v!h_F!@HrHC#QkiBFQl$U?Rn)z!j-aPe8*3V~DkiIga@YIiM)fZa6Ti0~G=DvB~
zPdbm^Q0-TVy<dwLOEc{l2c%r~%3rbwE00FOS@6;LwZB<ox1VZfJ9j4CB$QSh;1~~3
zGhPJJcLYL}^h4kI!z=iw3~8m2eE69UzR6s8c$_^_u8G&<PnTz8NYgO^pxZzjp*kIt
zQbT)oa`o}e&>*qAoF!x~WwsB(s5&qb0KUrm*U<JIEoCya^`!;@w>|}{x`n!EZ$(~5
zW(&OnLvSfxlbL34MU=cWzdd^cgD+=EWrt36AK9}l<T69ClR?HVH-SSX@4t85(x7Ty
z^T7><Ht*!+S)I#p`~)DrrC1a!08^U=Ds)x6>#j9mLAL+ui5|P`Do}_szmjl2hbra7
zlWAK{aZlW6yeUQSatO-F9pO8#p7s-bU)*9@q=xv^DDm-sg8!I~udC?rM<KU<+dKB(
zY~H+pI*KS@#8VcDg%9)FH0z}D((b<WIz9>8I{I%ZBOX^7u_<7n6YfK<VTM4ohXI7J
z-(;BJTv0fI2C|ph9lC4%Mzlebp7`6lobX}1+Z(x6S6;Hmxanp{Lc`X`gvkuS4oPJH
zqeK$>6%Ys!<0QwB{aSDcE&$St7$$|6Rype}&L5Tz<SB^Z$n5*5O4y+s>~tGBR;0au
z6d?d?jo3FSx@l0rpdt=L)=-L9xrGhF2M*;}7qRFsFAGoUl}X*iNLkJ(!Y9oQL8mg(
zXM+cxv|wE)6_&?i(t_?;v%wz+7WwZqhSkF$QbNJ{r+4z?r=+}La~{bSj`Ome@9D>2
z4W4jeh(q<j7CLLc;x-5iV$C)Sao=3l?&U^50b4Sn8^A;eE%IX~@~1}f916Dna)FuG
zuwLrYVJWwrFFPw~DlG``JXnBDNPWevnn<Y#04F^Pysii=NXAnD=qubQ^Lbjxz8wj!
z_FQg~i?dpLW3#`puF)i8Oh}mthTC(9Z4vP~{$6fZ?`!{N4}A!rH!w%I=W#Y6WhzlE
z>U^84cX|1zhxuV2005q!G=FkSkbSL(j~Ks{-Q3q0MnUl3#FIO9DT}0R{~(q1=BckC
z1X8{FK`rCl`u>H0*0k+h*m(h;viqf=9hn=ny)BA_JFvTH^i6ns$>v&b*#77)d0r(d
z3hoF`!?UV>JSe7%A$~98E;k*#xHVWtBA0M(a)kfaJhi!c`*{408EE!cZ&~Ll<fgf&
zPIU&2e?Dn<c#(X`as7<b^R9@E&q0UVe!JM_V;+J602MZTp%NYCI{dI>f>%4L_qU8B
zh`Lc!YA=62s?|%ydwDG{B>O3ZW<mgHFk3md-DKvS&PE<>AuFPeB_%5CiF;&;xeR;2
z+}xg>$5eNG;LgtWpU(w!Bq;za=qne4uZ>2<*kuQNsj4tTW$)>Is~Tig+;2R;%bzk~
z?Jynv+nsk(3?_eaCcl`TpD;0;oREs%^R*^OoCddFG&~!BrbJR#QPGRukr|#XZwxA2
zeY>}{nG{Q2{CnQIgnUEePm6-Y-)r8hmgn#M$s$cOzCZsuzDR9<;|(TnuId#vfc&zj
zz7LG*SYhTuuN_~D28wKc=EEuKrk2a<gVl8J-824FS~hMk^4J`kew)Zx{&p>7eB9*r
zgWLv#0KlCR5L1}i;Ile-`Wc%<5IiNC-Sx7R+j4dM4`W+-Pu&$a04H>wd+&YL)d{$J
z^!j#DdQ*zTjVq6?nt$hC%|Gu~b$hV;@VMP(pLx_CB9#$1%o`1Zce)OrdDTCOjw@2%
z|7Vfdcjg~1a7rrW)q)wp5rf2u<$Zi{=Umin+Bb&~x1}|q<;j~D4HEOAqZU879VglL
z8{d83pzbBzU~4Xldgh<P9n+oa{<~X%chwvFuM4plIG>PzL-~cbvl*~}Uddc@(xtu5
zR<1vFGU9l*T6M6U#J~SMqoqWw`MAcTHtV)n%f0$1XIY;8lt>nGs9r1O_}7byE3*Ib
zFh^xe!(Ok76S?eY`ug1!nN`s!^WCEnKaq&5!+QZ6<J|KRFmCsS(Lc*SPM?fp`oEau
z_bcyrqMWvF4_?kWQ(@yD%WZD>vYB>RwZGkPST;AJza5%*7|sg)^j2URUr&@4@N(=?
zuYeZ}%!nER-`%4!rEc@;My@_WP9y5)%T2%i^>&S~%2a&&yP?jEqj#0ap#ghcPMZoF
z{pET?THi0~C`#i(;AyPUF>g8+p7ShLB^-DK$VKzP-&dM2qTk$2?bJJKV&QgYm6Esl
z{N}^kHlHMR>w2lw#&*-lQ?6+t&Z;Rry06ulwDywFj)sf?#Nr-(=y7}m6nB_TEGScM
zMh5QGHvA&|y>i{bHQQVx+h%I{$9=6hg6yS|qT`l&t?8z(Z&h|0Z9d@+y@j4A=VuiQ
zdAH%U+<%KKh9STsJ6DO#_HHLLT9~^~Lf)JA+zKi15fO}u|Gzw~U;zRu5V(KOOaDKg
aBz`T+mZypZ)G6@L!05DzL7ARy`2PaaqOm9d
diff --git a/public/07-basic_statistics_files/figure-html/MoransI-1.png b/public/07-basic_statistics_files/figure-html/MoransI-1.png
index 768b926e0decce30295066ae3409d20fe673fd72..990ce7f362312ea273f93f8f61522286368843b9 100644
GIT binary patch
literal 15732
zcmeHu2UJtp+V)Q9A_nY$K-3vghh_m3Bt!+VAy!ZzfJ&3DQX~)(L`4S?97|M$#Ie9g
zQ;=dPiGzv=2o548KtNDx2rYz=_MZe$nE%e)nY-@#?ppWzvJMG3yS?x8KJT;lJ}2kW
z-aS@Iv(#n*0F-uYx7ZJWEcCAwO<oH6c-gvTC-l*|clSX{=o0`F0NjCz2{3`)<AHlF
z^d0&lJ|TX&o47+qya{wZp6(7+HGvMvS$7~V5x?Ty<Dnzh9jcHE9in&>5N~1<?+)VK
zp?67PJoL(qkEe^@a&zNzp?^g2bb35pl<N+1-A!`ip~GEV2o+2Ia&zgqbm$P}ibORv
zHS~BBdb~S57dqnU6J=0qaTVw#f@(o;bdd<^PDGCv(Q_r=pr+z;Th7^$AV(DfwjVhS
zfJU_V4;J@%&<y}2*kQ5x;F-J3X58_4r8hfIu2L>UpJ^(37oZUKd|kG>`>_j&8{&hI
zrnK3wku9R$BRQG{lwAnj5Yp*`R8}V`?6`VjM#cel_Cs6m5hZ{;(Sd_T(vk+?;xYix
z^I-t|EDwM(1T|qL8~_UyXv`m>D*OI-Rw3M~(!T%RlfSE<N=5Ca-e8YrMrk@BR%H6w
zwdeDwC_`%YrqS|ljuj)X7#Cl($6g<iCAw{%`Z~nB{{CiFd9bEVkq{EM4Izu7_1;lh
znOBk-Cs-^+F2Ah}p6J8T6+!UU9`1<TSvi`I^8L{{hs{UZ6$ySGNq|-?Zmd9Pcs+L$
z&?EBj?daaph;Udnvtk~oUj#$C?*e^+vta>i>@R!#6oE?tT{1ABElpT_3B2*?mqz2Q
zqE0D;WAXsrN}2d~G(xGh+4*tmDyxwGqFW-)cp3YBSMsZ>LcEipBKC8fj_7lH_wkl9
zLnnmw*|A;6iX#Ol!-TQF=x6p<B&9v)xjt7RRi@gwGxB-I2KUx3-PG{R&bYTO?rAa1
z=n761*j1%U6^Yh_zPxti)p%XcR`0vlJ039C5U`u9lhXINVZS6KuQ>fBJ9=YkP<wZ>
zBY5EaxX#<gf?1$`C8TEQGaJO&<=|zXDoBl4mF)Mf_(RJDl)1g=4fa(P=TgIRL#fu!
z<IbX|5&XeV<D*<5*NVCNU8+@T^xHWP4^a;QeqH%%SU4}3?ZZCKS$D4nwdyvgHzDBQ
z<RCA54^-H#RC{BFE8Hcxiy;uE2Q9vRsa_kLijFrmlfMF#vz>wC$B#pwywiQVeiklI
z?bQBAJbLA|$W)17I;?L;-`mf4H(F)dn;9X}>Ww<=i!r=i+BF~Gy)SDwzgTOd;<jg}
zDab{*!~su4YAEF75#Fb<;M4rr=7LeJ=a=f2i-$6s7|ITO2p+7R>+iRf*d3>16jzr0
zn&=ICB7EU0I^2K7qY~c`F+y<^=)tO&ID3Dn)P8?j`{4p$D*MC~tWAD&a-^Z8DWIjU
z=SCfVj;r>u5YRDaDITl51crR{w4?J5%z%EXm{->=;tid8&I;e<{NvqB&Pn#mLHafJ
z@e1SA1Q8`-=g+F3wyILI>r29C#P_^TK^$X+D0i$OA(I%_(zwUy14Crl%DtaA$Av#q
zleYs`{R8*fZ*vH~QuP|3;lQ;exKV>n?R@MCA8yC)<-i{A+SGX1S8T}9tyM@Ph}KT9
zfPD;qKmOS8r}M?5MhC;xTyt$Ij4yHZe~NM5*$aDehLX=#WMS=oG1Iv2b0@u%FvKp9
zrtzac%EdF5)eVUqc946r4h$tfihvwWBXKO5U$hK*qi+cYZhDzYA;aX=v9eE9z_C#T
z;biarD_@MBVUK+|#YM8My$7{&z+q<jHFHr!r!Dj3N7!C`M)=?c;&V{Hm%0m7XyJ;}
z)0508hr-6TRq}V7kfn{n`h<$eFs~jfl39gyXOTOO+1tk(j(THDR|O4Q>#NsYjD<*I
z2XydVuj8n-dA~v9H7>O)@ErQgweyC>hwdk<f|$xjLI?b4;w<%^5P8C@<r$(A0_~5~
z@A9e~vHS19p5Q<jMz#@U$Ua`K`Q_2DQ?{O+2xDgF>j7L5Uy;*?*E#QB!U)UeH#Gk0
z-<Mskh^W!e)VFeMCNEtRDvHaiE89yw^R)P$o0Q8|!5r9=COsm<lX|&PK5fbs4bc6|
z2^7A@MC?k?TR&aRrGnlS>A63Eq39>~Je1WT^`Mrcsi?>Ltb~})O|uDX&1`;E;BtG>
zUS23a)h@0MAJZX4j;P#e1RB|hGui_;fhYgwU9(kmIN6L)1SjLXa^+}Im7PU?q0B0`
z@#f$hDR}FPM>SEckZgq&tN08(79QNkv%kI-PR`Y(T9YoCTmf&AuD3`bL%gsG1a1`n
z8J|=+@1zUp2%Cr_Rbzx`muHt6TAbiGyo$_@({G6i>nvpmvj^4C$sqsZ0W=rX2T>D0
z?%22q3{_D)Y^;9(ccO2%G<%j55x!8ztGsma>m7Wzg0w3k<+r^!vyVZ763E*`K!`dA
zeb&^qJkH@-wvvJi<qI3hU-W-1K3uyA35-5aw}C!<NlwRuktKk8AiLYLmEZk_DqmRW
z(ph)y^(D_kbSVAa9c-K+0V9USayAqCQ=dh+aNNjj>(n4s&vA7x!AXi7jjb=`D?@vP
ze}3J3RWqDiysu7g8^(OQ@PiMR-s{loVeBYN;U|Z9d8~6zkJuI8H|YK~@+g=AW42#y
z1ev`E{U!ErrSW$;E%LcdYkO>#e#k&D=L7W~XKnC>_vI7f?jXX(DH)N~_F#!J7;e+s
zV?+7ux51Eoze>!jB(Dg^p$lG>z*ED&U)hK)fuWDRi1rDWR(pO4OEW#>a`r1b2Hw#R
z?%*dgSRC}j#r5E|jh(U5hVLLa7@8pFc$FOO*awa|E`fiwKwu$!5U5CawU2t3l;jAW
z?Xf-u<)q_X&Z1a{fQ~)C_G~sj<XvQ*A5{2!DpgKbSGF!Recf{fh3FAo<-f^Ic=6Eg
zmk;S2Mv<7dv9P`7qe`&q@HoeIoHrycx}Y&Rc@W8+5SNga*?zc#mGmunk82skyXGxT
zOMSXjUv2{z8(5aDuV>-+@)~OKtU`^E5Yh1*m520GxXk=)id$eQl;QXT;6og4D_HO6
z*@V5ZE^I$V1}%7()J{#yEejvXu(8I9<(WE56wNl@t*a@pZKtZ&_AHX>h%*k8=O%6T
zktW<S%Y;*~P7YifqIn@-{swo>P{wGaPOn@=gu#yE4o<l1<zwCB8KxL1q%*#K*Ylj?
zjQODVcqRg?8?xU-mNxz?FM;ZqX7m9&@@0>8dxZz$be2iJ9LZP)-K){i;8j-Z($HS1
zwWZiYSq;iXRyjITcyhF1g>ipz3iHvhjRyZ{XnAHTwaWab!Jz%eMz_{lE%Pfj!4KtP
z2O5&MJ|Z3sRv`EYD(i;Q;&8`-j=AfpyR#`yPlEPSWNCT&`u_T}Tp@|{DmJVNA2nc9
zMRb&v`X@+{gSGPg)l(nh;!RtuqvM&X7TbV}L74Lq1Wg7;(6ml!-Y5tS5_|^(5Y%L=
z7)3>IV&`x$5~&9VIQ#JRM}rK<_-{|P%#nU+3CK3-RS9jXlENad!n})lQSe9=&}DZj
zn#=1Gp<SfG_-Ma=cIu1L5K{OEYCfPV|L<N*{=p*qqUX!sUJi7vMJ|H_*?-~VAFkGE
zYgD-dnnr}9Px9lUYesDVczEojR^0n85f%v?0~(JK^h`CS33K<D&aQSVjRJdrJrNC~
z^qo-#J2x;5&_oRc9IcS7A={PJz8S1+dsYH4!JFj4_jiLAlgqEj!N})o=12uL3?jkx
zsm=hw>1lqP91M`{bo!xIpld-459$zLUwuMpO8-&uzdYWb8V}H!hnpb3K=ls`IgpiL
z<i}Sy$N^9U!GM7OQSqPoH*>tm7X>~;ZB@g=!wgio8f7nc0-VLU<t^q~>CItgp6Np>
z5-<&1i5N4rhknAfE;&=DT!@&Yg9*aKn&<Rd{AIjahT~NGjPhdCUB5AXFXuDHO6OP+
zT4_#D5hzmgJL&u^m$!PZ=4{){a50)nx%t{v+wOmvha%p8&v8@=G!VNb>(Er=%yFpV
zbQE7Gf_86MORz-Z?}hMLo=t^4*OCbt=OI{9+ZJ+;c;{t0EzCDOTm;z#u!c>agyCzg
z3C214j@BH}M=hCpt%*QO1vk9vOLtB-C)q_LQs-SxMQ27zC=<D~ppUtu-B!6hS<3c@
z$=XrD4IPKf;ID1Yz{cz+VWE_-U{yIMCMBUZ;w$^{m_MYROO4w#^C~>gbN?@{6P0Gy
zxjCbB^7tXchzqKWPB()0QVHo?#=>CMP8pd}pN3sB#F0*IlWI+C<l;A&;T$cY?X!{u
zB~3uA@x=?9Y=VA}_mjm=o%HBcryVkmYIjnOUJkLts|##LExt_O?5rpz@K!nWkg5UQ
zT-p7D(T5ir+G1pn$pHr&4RNV#CM?Zds&#lro?T=EcJ8E^YM}NV{zzC`UavPui<F)~
zZmTTqLoEJ`Ag`7LYKbNonU`wYCvwan08~^aac=WZ{>H|AYXCvrAxJIN?gc4x@3mXO
zvt_6H7eyM>%s(tkt28#nqh3sP<5K5#4bRFKWXq~eKrP?w_I31o1(yxuCsGs{NPG%)
z1@Y#*Q<IOGg4m-i`hK}d(qfP9Ei!hnIV;AnIwm1qTb#T%%F>bBVhgtP;R1Zjg2W9~
zfFz;B%&A=tCp%C29l9>x8!x-b8#y5qw0O+Ta3p`}VJU?P36W)HF-fpO^{*A4V1h%7
zv?IS(xWLRR%k~y1(58Gc2q5B6W&&b~$Z)C=x=Zi4(szI8AsAUb^2xn-hgfYEU{7YV
zp{k&0R#Cn<Jb%?^vCRFGal~RbiKIjh&h@YTh(0@!3e}j`Qw}Io%D;NIB}Dc2PW)}l
zK-ZzJSCt>HiFItsQL%3R={QYrS`0@a<a<<j{S+rG!pEMdNNmdk0jpAlAFhoJ99=%a
z48q*OwvugAFn=nEj-7JoWil1bv>mmODxC@jYL>Qk9XoZr;LCvkT5flYF2Z7yMw
zTdEEaMiOpa))livMGNIKh1{}#XdWDW>2F$g{USokXbS6T1AH4+45H=hg-|~HZ7;~*
zfcy7Zw)YQQmn7yNg7+7__)F3JBh3EDvOf%pn9zTV0{_U1ozr;sZ?o+C|4Jg)w#Iy0
ze3-N{vL^D-OFyg3Utl)Q>%5;6mRTS?J(+<_%Tab)mtSYPNpH{ll|jSAD{`d^_gX<B
zf5-^T84oJ@JoErdHcdXAmnvE;`oL_hhxdr}^Q}%+$gp!$DW}D;_i8Fqm6~GSzh1eR
zS5))jsmurQlHmr%^_pE?r|LjQds|%i`^Ns%?PsdHC)!J(x5ngF5Y%nvJP%LC5<Kqj
z8&)8dzwhGdAxb4$DGs><rLR2o8h2No{+u#GcP1~rGwhTz9UG7qX)=*)$HR!aYhA)=
zB$zas1x*3skfvis{eo0Q`Ry`yvy%z-qKlGcjfA-MO>Qhr{UmUwC?Pcrem5tW*W(uH
zkT~axA>^_|dwjaGUa^ERoe?IirW;k-HPR{0JFx6D?<?mpQp>gXs6X9x1Wx10BawA(
z<?0*K81e?q6$O|V&d+aIzUi*&yy4u*lI{&$;Q7I@HAgTHNQj%zqOgg=T_&*SfU(ih
z2T$>_q~J~Yx4dcoRtHz_0%x>f($E!^(1xlJ??k5}%&o%5R%nj*rL!3@I4vcEV<Gjh
zTn7C`Q(0>cj$g(bwo6V<wt45QrHtI12c~K$7QH#^2a_i3!>}&N?p%buD$@*?2lpv0
z^OUapl`Ko(Q+resGDQLsqHx|`=GC`63xlbq<$77ErNc21#i%8*B5b0S{_wmPQ%&{q
z-slmH`$-s;M&@q<%6@Z|%fR4qui1Na_u7~=tRibn*taS{Umx*Y;pLX6B>8P%aV<N^
z(@q?R-Nf&W2RE9wu24SxIbzvjn9ISsu%9uPAT4gh+-!M$dc2di)E~yP9nwCl0=5Me
zX*b>w3p;O5YvXHNy*09Uf#8zMgh#>RnQE~Ppx}|a`};kj4{9!%)?JnjtTrNuHNJ~C
z3uZvOl5+4S7(l`0-vl;TGT)MVek7M<Na>DEImO%w&2(40vl?ne9H;QHfwrn1oTFT^
ztmD8;JL=p>*Ag~pjT;e8ef2s+OiHf9<}_Ao0Baxtwt#=h#mV&N{YTgU%=V$EHj(h=
zW@duHS2_JMrP^oNMax&;_#O~mC{f;wSv*M!v3Qsz4Gi+{Em9@Pf9#Z43S?&nOON`-
zJrn2%&M*<@Hy*6kB~0NR>?GUBo|A-|MUhdKz@@gb`9hTx667Mm<sPifvmeBhw9r=h
zet|juFxL}*P;qk|m<H33ag^!6I+yW<fl3n{OyC`6+7;z5s?H{RPYxw<vH*o>b;)*k
zn-hfXnpx72Opqs}PY`4ROT0=x_sEs#e5?f+neezU4#0;<a@FNGNVTbD(YRX``>PGn
z)3~AJFRgCQ4G@&R!|8=<&$Kg3AI@QKd(tb^RhhCBa+M6*i<2J+N|K)DKV}8cWb9}*
z+XON~E%!~sO4iAhWMd=xJb>2-RGffk{a#9Sj850WNh`>K_ZA(Q*1||POTV*|iW@f?
zESTr{{j_>q%hAzMeh(=V;a|K4MW&D)rh}S=89f!sO&(6%95t1$1#_V_Oa~=x+~QV(
zC~&P3p{J;jQGoj{;KNJ^v=q33<K_4u8Bq@FOMvc75Y}RXjd&JMLcu}SXEyK5li$Ul
zN-8CZ<19@Zse(~E315{$GPDoh;)x{~3G(lXlEz3)OCThmEdNFoSsPf8)wGcCV`vCl
zSip-In(ma05RbSLV>pCEFav?eQMS=yqBc43wxunGXCARnhStH@R;ZE|JH^pcF*@Ac
z27WT{w@g)Su(TB@J)u>$Hcm+jq(%i;r)ET8I4g(we!cWJozfdwrudEAm@dnOQ^A|y
zc9)(aMRq#zLbV4NHV4dy_O#BQ#`7!=eCJ7hvh0MuNgz=M&C?~OM!ahjjEAp}7;n2T
z*rz)aruz48B%5tEzH`3Q?oykUTdy<p3I+-nS<GZI24f_cN1$&uo|aR`?TCYyTzscs
z%qw*U(rBp(6C;0PMko>D&iO2pgaV>vIy7P58@E{OB<1;)s?Gk=?W@TvrX(7=({Q{+
zuKHU5z?pClX8$}Y2R09+YwY_)2L@(Ohi@Xa1Bo&-2mr11XR>Td$dQ@eRA`?SgeDCq
zl+a(yBvs#Fq(Ow1;=HMhlnf9Bx~@!wUZNS0GB_n(-e%4Zj}XSyh>A)NpslCT?@ds{
z=dv#0u>$Ji*!~1fn`JZ5-S(*brz*lm_1K8Do9-U`YL%F5!M0QpeZ5k?M&MTr-JGQ}
zidxIH&*n_TRc)mPtXg9VVYdu8))wC=`pHHbx*#AHM+oMMWd$ji5j;C!J=O(_4w^Q1
zmOIO=SEHr~86C_^omx<+PD|!x5LI4`a1pjJOou}Nb&dI>8~<PG_qztkU1Vu+L@6`G
zQ=QTW-AtXGf3oT{4^c3kvd|7|Ca&lDvu37t0<`4m6OyjQJ(8=LKLhRM3^#xODsxzC
zZ%IO-eA8r#Nv2vKVv%EnbVXe>oAXRW0bq%?__qM+CAWW<FuhqZ-TnAs(w;}Kz0C{H
z8lN?smKMiLOe0Bq@Z8LPxvMRU-`G!!K?9^WC}np|P+M@;KL`JE`!wau3Xw_|YZ1_r
zubG_w$${$T2p1I)u8CHkiiyfp3PPd!raa_8?xIMoG<fUTMUf6u3Za+%MsE=dmI^12
zdjdTv-)T7rvE;G0s8s(kxW%VmG@c6o1l^&FL0VTn<}a{NoUR6#nF1u4vYy^6&q%W4
zNNF&_e8bi&{gb5FO)M(p{wmy%H8XTR;mZ$8&unVcGHcn{nel9(g0Iu1(sjr+N0u;N
zewxhrT`e+GZRT^Mxc8n-TeRs7&#z|Kh}J%C53W|4QW7Fg<%|XJ**c}Jw;>(S4kCi>
zQ&})wG=DeaNr+O7`Q7aQdx=O1IYLfASIpFndS&2@3J=^kEiWa#g#&4>19C<h%YIEm
z$x2V2&_7MEH-o@wGma9RY&s=y%crKM^?ObJ)@jN%a2b!v<YTk=_ybhjDbxI0Q}c7=
zZ-#JWG^DiC`13pSdN4-Z;Z4!W1n2^pc&#iU$3M@}dj&c(oQ5uq%uKYu)y41P^!t^`
z7X7TRx)b@^c_-It%JpZf{(r}G@K5RT&u5fB>q2&?@z}D2@UdY&pPz>y%^Ay^=9_x(
z)gnB~zJNpNHAj5Qm*)7GzKNZxA3}L*eE>lSTh#D*M@~0mV`2eDP&$^pCvLLTWUybK
zRuRxFWVZ#D;P>%<EjWP;=Y$cX<tYmlpr}I&^0YRJ;OCrXyQ>4_@~zUC7-@7_P?V8|
z!MPU{+>p<DFI3y<jcPc?XD>3@d!L_G;Hp$ju<weLrFvTVBWizM5~~d7n?UwPi;E`q
zbI^jLjCoPdO#!f$69`L5a5dx1@s3qwZ$ImulPB9KeQ7y#JAK8bo+0JUb%drXUHd0C
zE{ikDQKa4TTCJIib~OSj+v<hw%<?DYdXK#ymrh-CPpdgZ-Z!Iztn!wm^({x&MKm&D
zkZ3*Ec>hob){2u>Ht-}`JJl|C;Xmlr5U*ILl>`#4TS!zBGc(*N;W&Y}Sq>w0%Jk=5
zlC6z&&*=T<x-k<bLOYfex&k55;oANqaIpBUb)JoIlZbhogI^e^`*y->mw=KW9uZNl
zD{AIP9evFSRT;Z8S-)w31HV(~=q&evIqV`&`10`qv|j_V#^K=ZAyW||IXNy>zIEP<
zj2y#N6MLuWSLsnIefHvh*x=%+B`X_SVF9_tFvp;E8=wLc$pGTbxy#op_NTq~Pwud1
zYk)Ej+CZd1UDKWGZ|8y*D*#<H2$8rb+0oNl4GC-y5RO_3CdU@<Vq}#FUY*pmqJeU6
zFs<dxYxnuHx~2`@3T`EsvDR_h&v9Um(+sG0yT-}XB&A7_cb@apQdknC^hOL;trRdY
z;LH8G`+$ooFlO}exfazqEAy69u8Vh$4MDsPhN+)O8tpA{1w<(}GdI`${52F5dQ^tx
zL>QqZU|^mcCT7AXu-LK74?$V@IoN_I*>zSoYfi?eRtov{B-v9UWAStq%p@HDx$lqV
zfD7`k*{%Rw0`3)<b=hB^Fr3!9B_wPbVCVZsCzEs+m=2b$8TfgL+>Z+-3$Smrhc1)P
zNRa=z?`c?w^?%KFaegfNzd-*ru>LR5|3w+iR?k?(D!xFj#w-f7mdNEq{_Az!kqB3h
zQeJ}_s#$53M7vz9be`+Mz|#MTFMijdOMd72DP-@n8c#c*Rm8b}<ClGKr04K`oqx>_
zACB5$?m*+R>u(`+yoaUgAk7O-hb1#fUk%M_hP6IBh<V$?yeM9doPm~c(&%%P*>x#?
zd0*tPkO;~X<bzZ*_F)n%Wj>tiH-RI!U@mwctaIN7GrU)D1zg-qY*VYvIdTwf-IcK%
zEc@V=zb(jj;ex&MD82tm(*i^D>ZoeWHeiub9uV6T>>WXtgYGL%j>G)OM<bvccrfLA
zdL@RB>FL}^P3R`!q)~Ye!%}8{=s~kzC1M1+FRKDnde%DLr!?Qu>e}3;H)Zqh25!`W
zuDS%)O;t9EnU$i@95jhdM&-|NuPq@OlV*5lrx5N`W2`1^{yR<omIoMZBX89;=aV8P
z{Urjz^XeKK$*`CnH4W$&H@YJu^CbQeAtKH`H(9lJKU6jHTh+PHLSt;i2E`AJh!E1F
zZRGe4k}e%iVc=-`lNymQl<5^ExjQuNe0&X_opSO;L!~pt3`ktpg_4ZyOY5d?`o&1_
z_0~|giX`xUWkCwSx5Qdh7hR(yvScc<9z_l`voM-xCBeMuNCXCRM|{Je8kXJ#nakQX
z6Qhw50)1ciJ=i_+OrW(7>`sc^R;eAXXLX-d;HYFFadlV5E^w>MH#p4vNwHSIiSgcu
zUT#uTomcd48(G+xLtq2Ej&WN(%#(r0w}75kpk`3l8IM+rt3pe7KPPqrdleGBu&R<q
z3%~?_Y-FYqW&D(E5i1p7K&m9Dq4U}ac;PR&HZO`OPq-4qZv`|O0TaB>%*_xb;1&+`
z@)KPWGvG3TDiDgU#3Hj3g@iQ`CN75L?;yx$H@aQ3W6Li#Z<BFxu@RqWhE{u>nR{$(
zNO!639Jx^|Km!=rKYxURaPf>5fPQp^wt#y0OCQmTVH<X~gjpHfjq5LTSWLTD7!v2Q
zL96+ap-PFaYDw*DZr7f)ON;%=7H1f}vkEz8{W4(J>wuh-D9SwVWfvUQUPz4{exG=F
z-kPSL);Rh?oxHqrvF+yNHm~;ZK<$qfv-4bM+o^ZnH+{O|SpN6N^Be!eUH@kuL+R`8
z;g)%>FXp(H<WW)l{*an>)ry;u{Tt-0cDOrr8X-c4dfs`irz~Apz|yoU!3~x1Zyj3-
zMu&qNx8E3a&SoTSQeTj@&quZE%ISdXu3v)uh69KU)!6>9CG`eSYm=uj{(_E^E+>^i
z*z+an9cu2Uj8?neuU}iO(b>APtfIH+OWVidb(GJi0|fL<O|?Sj{p+;Of%?+A==<+0
zyBSwmT5iVYj+icZI&%yUMySA7m_De#r?f6xH#=;Nh8w%D3cs7mE*i;3OVdz#H+}c}
zDj@DKaK^l?*%xDvm80`WQSINY%RHSB9rk>@h&pb&6c6{&>>p8#xt?&rVpV9!Vz>*U
zH-bCLeZW;OsU0C+d|zZ`=`-I!mcm}LWYz7^MEcsMZ1%dPv$9PEEVnn9i{^3`pzr9o
z_pe!aFHFu}fg$%<{36A?*@3+qH|$?KY?8;A|1qK`pK08*4h{+&KMGg-8CJX+6DIG^
zSlF<(I=!=XyBLCq1BZ8mKEtMN&+^-?TOp+K)6CA-k6#<ym194vVP)?0{+w<OsnIk`
z5(`aMJ-C5@e!+dO1x0MILn}(^aW-y1iN~#*e{=Y8*Sof)$`-kS4<qDXp^zc7_bCIV
z^U!#x1BZ)-?_D+rqqCu#)G=p@dtIbynyyB(;N*MpkeA12?ZS4_GurkDI}=N5&lkHD
zemc1l!e*_y46J(Zo5FWY!>0X7pfUS+Y79=18`EhZO_Qyq%Xb<paxFdXBZ1ni+jY+q
zONv%^`O<j7J{vPbms|k#WKB?zslC$me(P%Tg)I<O>H<<7lACqhpYekGESFL%mt2G@
zsQ}ZAm3mMm+=Q`L+MEuCz;(O017@bV*Udc;V#qmRvml4hXa{Uv3VDmm*S9*)JQo(M
z1A}CZA(tE<do~P~ga7#)<cvvE%uh)|Qp4J&@fy@a{>CsQ+(pZjb?(U9Z7-V92OdIr
z+-%!}M+h@gs^Gj_-~1Z%VfPZtjdewuVjd0!cdN6`JHOrb)QxOf^+M$*$g|24-68v1
zl_iOoe)}S+zMjw%HsL{*pzk5OVN2|S0~O-Bb4l{HJ%#Ulm-N3%$+)glzp20uQOEcx
zLLI=KB%7;|^z8e@)hhjOpv8GMJ5Q)^Nk)ctvmp`JSh9P^B9sOJ)YsGwpI8^m(j{r@
zhKO_Kb>5Fih``*o2;5<J;FHiC5|_NZXrJvmEdl`ln;us4TEWsJUEX$I-Msh+p4s#2
z5h)4pG2}DS%#y=t$zKM9TY0tp<{ZH?_zHb1#1Qmr_IHnzBI-Di)$+je`I1^4BW8D<
zku+_{G{I;L!;ssu93VgUmb~x!5@S;%;<VL{*hI2YMR``hC##o-_LK$>9CeHZzf{#o
zkNq-kE-ijSF#eR0nkZ{L?u1&?V<EH}fgT*({}Z?+R0!}EJf`)<MM4jd?G~}d`uiG_
z*dex6gElR;4C-J<`vsN_zIAkj!3e9-nJsMr&~GhycB5F}n3kzLex2TGo~gGKJD{r$
zNLrQD^};f&D>?)+;GpT4Ag+w)1$)vW@-Pkej>>e}qM42_)BwF&DS~@23rm15_leUI
z4V*TUzFE~_N9X8hw3yo$+BW4^oF0$!FHRD6)@JiL)I@77UiSzuZegY#xTgX=;DcRW
z-@K^2Jn|-e?6Vg=NqEnIUahXqZ}@cMZ0!Zcm~j2(FH~-(5TUFJ1BTZ^;%?g1Y`SfQ
z=GiB8%O=rV^*^#Sg*W--uQ+uIr=DJ(g@k^R`(q0gTZsK7Zvk)?@_A(|)d6w1GfoKI
zPu{u$SwDKxLLQ`0uqb<Tb~g0PQ0yn$yi0F3l~Tv(M}=xzdRHw=8hf#3DIlkEcjK9>
ztCU~{p*I}mc7oP?6|nczO_WvkSk?F;^!@j-1?ECNiCYI&vyzfo(nt_cC=XUX->zR(
zxLt}&d3#!@#=5X5a#1mVJX~PA#z~4cxD1e8xeuzE0_jPD*Twa(iK307ohyLXnkU;Y
z#@fLuS!-d?g8=BV!23?ZpG3Z+ieb0}XE|DJ3Evxdq}s&{!wi~77sUm9q-4#%7!3&5
z8_EkgPI&ub4HxeI{yuDVcBn~(b2molYrxyXTw7i1aNa+MN)6W@==hu>B+4p7&ztJ{
zTLld#pZKy!#J?V4U#|mhGEO+jQVMsba0QK(XUC+#pg1sv&K^0Xq9EbJo=f){t_V&A
z-4<nY%b%CC()6Tox%@n1JohG7*r$i?6Kz;14LS<(C~Bu@T&d_<&*LOsjc7On8#NHC
zO|sSR)GN-8;Bo}R*T!+@>=`7SCap`4Bu6-o{;z!IHARH*+)_$?@M0GeQbw@Da*svE
I7MHXC50;M4h5!Hn
literal 17446
zcmeHv2{@GN-~T<MNe$sdh%DuVQfNWenKsL589EgeMiFfcmF#9prIIb36fsJrjx14#
zu_YmCEZMS*E&IL?GxNO94B6`XE${VS|MPqQ@0sf{#=U(%-_LismuDU~4fpFULyMsS
z0Lu*YcO3v=G5nD$YY7+p>wUwH-{4;@hI<e0hJOQ~0f05o&;T0ndjPObg8zp9U>C4o
z)*9CE6`%o^2h>=@O*P<)Q)&&^HSAY_bpU)NS;GyI;0p@S00GwUr4azvu`A(MQb0fr
z`z<LcAPN2e1=Q38)IdqrAjw)IDFD8#*_CiL=O-ztCaDI#pd<(?D=VuB(5MNpu1SKg
zfSRc~xHr2A{DR<C@LLT8!NWl{0Z>g6=P$S`yKE^-wi?!QslEP@3jj!Xvwsl&-v&+s
zAPo$5Z9nMzu&0h``Mch|R)=-loG<sM)_g9$`pC^%?bD)=ux$Yj(rH;6K1j15$Yvx~
z;l;rNb<wG9>N+&4Qr#<GrhVsOZId`J0w4t}u0Q~wDi6StTS!2P`1`}pNg!G1kogvS
zH7dP;SQ8e^-wg=RQI+~%ZkyMIFtmFaR9q<eh*jn)qGvy|BI=m>sZ)A%S#KOGx24Z1
z(N4^iUkIyQhi<6L$!2~;q=m4q9gt(JPs>BfKhQ#vb{c`~i|a_yHqA-`N~8|_iFgmB
zYT~2<DWTwZuuTXhus0ZaUvfhe%Kq&(NbSn}s%Kzy&=81t(P=iXSp!E=_CiA7wjs#A
zDo=XfxF4}fX;~~kT3QrMLbiu2uH;4ue99g}P$mmoO742qIFYF=H^=4)i~cgk*?LT<
zdy#K$@CH^(<iHqZV7#o`4afi3ry4Pu7gH_wA%l{to&Kuzb93%?#);7~y)s;_iw<-+
zXo<g5Sa~QW^T8_@U}uFF$|_GZD0Y*o%xLZymQ%%4FrsaFPuia1W}rewwz{US`Zm#`
zrWbVIeRJ!uO`Y@Lx<jB#>I{q0bM2J-CB1sHUy|~;9ALV*g|h~;?^?>0@!Lx6Bum9+
zSiNbsX?s7cX)BynSWaeHjxfE4@{*~asTeW;CW>nc<8VRd#|o;_dJ;5tUM4G<V2lgK
zUs5vCsM%>CgdJ6~*GHW5wLUvoaKDi3I4Elb%Gzm6maT&BiPe1TNC&TeahCQ&E@_QN
zrkv~kM$b#~d*ss^yn#SA(dNCN%)ob{?A=Bi3pU-INGux~8tsgvL9|0RzLk^*j7#oD
z^;5-(<PpIQ4uVD}<ZDm?5YH+@P(zF<Dv+~O)!9aQLm#7UOBaU%kKW)yORre1E_ovo
zTv2}x#G=N}`w%2|26y2%W3+T}YC5&@q;iqiFe#t1f!tPdEvt9m3o`xy2qk95WRA4@
zw!e6EZbepjW)=&nf7@EdV{@zD+qkX7!OqIdEjzlEU`2*Gidb6~QqKuz4d>H)X?ajz
zVPWpuAzNl_ezm96e!Fd+ZmAQJi8-AuG-EXbxLHfrGg^B~QIk)_XR4c2rY~y~^fjo`
zr!T|tVfQ%Qo3>PIviBI`I!?z!Tm>rK)A8t>2)0)#J`>!DJhF#;TV1@r<z9TQ?-p@X
z{K%dUMPV3QBI3!K?iA|?UzpPfWXh><H8oF$i>DrKbkmoVl~%6%Kk?nH#{Aj_$-@R^
zv^K>M`(yY2@u6t@S?|^gTWkkg=R~EWD1E6w1z6k)*n;uNpfvLIGu(%BpFLQKYQ)P&
zJbXGtY&^f#BlUHw9Ez~wuzNS6lxtn(^IA8(rW|iQKd>*vyMlocL&a+a>k(N!qeUOa
zj;z2AB@NQY-tJwI<;M}zr=wYMFIlSYocp9N`GMp|YeWl#oFsDT`#pQ0v+_b^t@vSo
zk1$V9;WhNKQDN-ncYbFVf0ZX$@U181iqwq{B)1x+y6|qq<$qh3I@(yRCyok8F|tSm
zt={{F@XP|6^)QDhlrrQOjrq#BRgSrnESIcvw+NXJq4-=y+ec4@RNvX;RFr;vq&WmS
zjAOn|M+MxDZw|Svik80b)w{^9{^>T#r1iiXU*AlJ&x(@HDXhlwoL_>KisP+^S$)Ze
zi;CU*PYhxA2(uM31U}8@2X6MH8L6Qaw5P==l156-)q5tcfVD~1y>x*kQ9(>Q0-c^j
zOG_NjQ{l3OL(W^9cepF3byVj8p{)pM61K?)GR53DjDPK+$tsW5i>B=^MHRb$`Yiz@
z5j}lukTz(GW_&@xqZYKqq>{ZM;+(%M9*j9r(@zgnkqbwtcF=2Kls&O>DzLKTuK1d(
z5Y~V7-0)i{^3<wALM7qGpb;o|v??7Oh*bX#@hPJK=L1!z3vA?(_L#P&5;yZ(dCWJv
z@Gw~k0`jN_ouQtROl0(As`hfpuP5zSLp<fNpmp_}TlUlg@r2-r^Rtcys+O^tsU?1u
z$Pi{3aUxL|AHq1TiMDv4w&g|7l;u9w^Ec!`o?Gia`9_`T$q*cm(sn{B@jtH)i>C%-
z5_cnDi%Fmx`bA8SiGK{_k=70|b3#Vt1!f3350Jv}IJL#EM#@IvI5ddCZz<MAEV@t?
zmeq&Xo*-nN^T=j-UsmOhiSfQ13-P?6dwk@Hn9RW@qb(}QruzI(UxE;B$L8$k(S$pa
z`&ymi2*>bg&Br5rFs^BviiUA#Gk(?IOqlhtb)@R!_K5?iW0{Cw!*9HyUwslIcV(mB
z?bdxueDl18**;8r3ar~&86poG`Tg;6#2|^Cr;R{jX0N+emtL=5$<{JC%ifU5)S~1)
z^#f(9JQfOs_3vV*U3g-d1xu@<fJ?nz+cZp6sa47Ya&TB;WsYuQR_xjSQY5RtU~o6r
zS0PvoS&8H_dJ*vnsu+Xw05i~)eskFHBibT-f__Hp1+q%4=;(yiCweWJKv|@BW=SC-
zwJXY28{^CtHytrY(?x3;?@cluQ<<&AmmUVaxeXOkPfK5G5cw5J5*VTlVrycbJgKNH
z-&TvTQTuKJvPzQrLCwJ5dLx|a>P5VcJ^ku_q&<dQ;ck9jV|HA=G(l%eZyLe9XE1Xt
zW7V(|Wa%OhBUM<G*fcu0Hj;<OB9BTt1d6(0H)uxrCii~Z%f~iET(>%-aPa<K<GwRn
zyMo}%qJxIK%cV@!pk6gKmRoa_iWG0a=g`{$8B>`oOAleDeYTq+=c8m!x%RPpkxQlX
z&U@u*dowl_e0W81R|swlDv$2HD~z3J>!T<n2YMJIUM8qPr3x<|2?=L)ULKRnG~K}e
z#d_qKZsI1o)u)opwM=Rn+M;V^bpbhP^i;f6{l%-pDqN@8Qy_+SxU;WpO!cipn@jW7
zu=K75exG1Q>O@WV#+CRu=8X=Kef7&7#v+IG2y!&{iQa%N{+tc#NG((v(<J=RXc#Kq
zIjErVD?XLZm3;lcQ9#R}P*fm90aD+uwWhTtXcG;W*)otvNaKkZ(6FanV!a;elK#WV
z_-cy^*GVDpr}}u{2-s@~wmmK>tFgf-Nns2S;Hx}XRl>M@&DXna=%O@M1Vyq8$y8}<
zDQQU5^ZQ=%)iraa9RhjL2yFY)w@rB9q9IlkMQT3kc8%`#5F#m~LEdPuQkk9?f1eiX
zlmrUPhXP=`5dZ>0e}DK139;7@QAalEE^as)0X}Zp3@VW*L%%>oim*KC{B7Q>H(117
zODhnILPxBU;j-Uo2)>D~B%QsCl+VitraVGeb=}23r}xz9h$2@l5?8n2>k}RseJ8c0
zvk;?8kFMlp{k{?<AmC%}7x-ue8k{j1hULT<0>hlg!IoC1t|3Wc!s)AIb}9&gFY)cL
zw4;we#P!`udHKa?6j0gVs;+zeusnEvyrgi`=GQwA*k#{`>i74!0@q|=U@?yXfU4x+
z(E!7e90AMuY2|-g)I8MDIteA-7%G)Yoa}!w%D*FS3SGHG^`wp0A8B#XtPR8P9*7bB
z78fhYR}scIFc9^~SO_%Wb3M+xXs8bp$VIgwjJ*CRR}5eJV;Fg7Pf#5H2X5BT2kv@b
zp2jp<G*fG8h-92%lHa7k+S<6$p<)GCt+=l`E43wfRSErXSMg@Fm0&BmP+;8HmV6s}
zo9XC=>$L2@3CoTI)9k$sQ9~?KlNu@5hlLmIY(C$*%DBC9{EF)pPC3%k#s<dbz0<dQ
zs3QeYzg5RxMF&duTu1tSs$#ug_lb!wafxKAoM)J};BCFj<|0!yXZ*T*IzH^#&PzVX
zINiQtYVbd0#M93l&g$Rf1n^SR?flOTvl!*o)z!^Il@D2!KQ$|0jU?i9iQ&-bt6=ig
za6>>Mj*J!r^ZUp^b=|(luaBjOH%vsZxylqShT%G3DN>)_Y)1fAxe`evC<yXSOw<cP
zM<8N(*etmsO#S?s)wY<R;42Ne*tqFJib*=CZOLtL-gx8e6}E}Yz*z$Chr;bMSVgkF
zx@EP{1gzs@4sW}t%W=2;X~4wr$dod!Df|{Al7NgW>_u5`cQoSoar99DBS>(_BXoFx
z4bK&)++2mnqWc3Toz?a&4g~S9&_>@tmG9;Z7<(S^X)j@HCDY;PgccA!C^E&}Etosv
zNc+V@AV^o<eUHUV#ROr=7Q4&45f3TVlOtzM49%vnOo&&vW6=IY;e#t-ACODORpq9#
z<DtXjonpmqm>PpV!Aq}TH2pYNa=Z@=mP9O_ylVE+2Ub0(UatEDJWvC9Sm(G%*JDo5
zRTjeLR5f4G)L_5i`GzSqgurw}8WXIH`PY%o-sXg?X_1Im7elgtR$aZM(d+%T@ewcv
zcC;RnIKt8<jakl+q?u4?r-pp>y&jI2_R~>##Ed%Y{kw;%t25s~Wt=hpCC~tRGPX3$
z1SarD{n?pfzpK>@z*3nREJ2bhdH(d_=<#qke+I!5y$@G9r<jn_-u#_@^^U+9&<6gu
zzZ>!U(0}869Q-fx8SIXDo@kzR_Qs=o(>>brT@i7suOFQvla7qvACULd46$<~hyUR5
zr3#Y%NYdhtt^l*gdl*+&`0F9CUc-+0iY043-`&cK1QFXOG^~Q#mt&6SHXDIrX%C5n
zJr{41<C_NL!Kr`@zUsB9?{Ovr&O%_TKHj@~xmlCkOJ^-K5PK50^U_Vl0l~vsk|;y2
z`2wF`o(4EA@=KOsA;<F5LeE${|Hko^X|XEGU(R#9Yf3OlO!Ul@(@crRF5`VO<z`dD
zBNl&G_~+vP*7tAw{_W5I!|{7as=7Ug+}75X=`sGLiJA~Gofp`_X<8SUPE<8{JDED{
zpE$3gD9}uM)k{tK({6EiVw(+T)7(I_@cn$u8k;T2vd|{Ri+0WV_342g9?ClXWHp_B
zJJTgJrupDTc$%4w|JeHc>6M)lT^OFCW;21YtyN@{S9a?%onWu*v9<n-Pi%Mmp3t6O
zlJNV!&QqV0%;drAHPiX<_@u@v=B2v?EmsXuJDtv;P*N=a)Zk!QO=aUaxx?ueIpL4~
z6J4*#wHxgdTaLxgCQN-{Y;;tQ-GsY_9ZAiX-_(~1#o*C{ja6$CzGu;2V{@Dm<@)*h
zGOAgoH+?#s)F<igO_kMKvTsSEenTt~oc7!R3ZGt_U6Dq^K%OFPa(baCbloyIe@@a&
z(@kn8iTl@3Znq7FZdI@_VZ_X1NpVYp(opkci%T${Gm+KV-Q7Jv?-Hh6MU~IV#Dg-S
zPFQ+ooxj(-;shDDWOsEaL9cHW56_=-GI@_=b(-FgIO$nmXK5<JSueiJU}|Pzr}-Yy
zY6gupRO5Mzoc$(C^NQKAS(_=UB;>wIcU_a!R`$+#klZAZ9lL4PZpmE_vb=)Yx}+48
zK>aV>8q-9u=Y|esm3tL++;VqZQxZdmB2(q~efC;?2;8hBjT^GUT=e44?wg+BxcZh{
zKm48WCF!d_Nk2NQaq)Yf<9UwTCNM$77er<s^mSrgmQdD+?6mb)yGWR8he&*B=5iaa
zyhpq5v?#6q`#?`(a|&~F9H03w+32<uS^;A`I%BL9J~_fS6^fd6S|;Tjv_=5N_-1-c
z_GGtQ=GbJ^q|ZGi<bdz&knGT@9nG&GP^2KujPZ3AJa9LaIo<A17ujhYj4JoRZs_;3
zFnc1o;_3Sd-lFC3EY9Zo&_)u0B4cv4B5%^O!sgUHcwS~tjfyUUhRy3JPyK7P{><n3
zlX&HbT>keiYjrP`ILCi(d8B14#r@T+1yzp8eko{m6J7~S;SAN8fO?1crF0nTE|wq-
z4|XU$XDzi3lio8u&j*WDx4lWT+gHELS$6PBC)uK6X~KJW>A<m5@D-Evpk}i)B9v!t
z$Q1mfv7HSHh+TqDo5(~>P0h)Hx8DYE2zf!O-1Ndkc~G%r21CT7)=9(@A?+8^^GYux
z1)k2Rf!#Tw@vxpNzTYxU6B#Rc07$gYfGCb1J05Y(=2A|}y4<?(UBWKGbL*O!zz;s=
z9J{8(BjP-#W*WL^EuEfInpZ)ibSV_<o`au5D)WTBEse{%>}0a|<uwaAaaV7TOlRbz
z>`i5$1zfg|;FP--9f6ln?3K)0k3R0=lD5P|;#nw4W?@gt(spN5@xFzID0M<QxpsdZ
z*!2SpPp~Evxv<zSPcZwsKW13+<gtY#-El%nb=<TFo2w8IlLtl<a`zptY~ZZyenmHw
z^f)@n1FkuR@NzHE>E?{>!LriI<1J>7XYBy5BSq&E+@--d@5!o(B1DeHEpp|Z-@T)&
zzbB?Tl9-{h&86%@#IyXgUC^S${Q+xXM_?})_xN>xYicCtxt}Z+{()ez>K2LP`&r*E
znu4IQ#p4-u$B<Gx7pi)RvbH5?0K<~Qfi6a@U)VyH6gMS^lsPR5U6IwIo$fa>YVqv{
zIqn!zK@V#PDGF&=IEfqnAR$AhiURiH#WfbXz7R=%#^2fg;ulKV)iL9j91qjH+^59T
za7FX7?Hs2JNa0gzqas3uvc{nyWkbO^27qUZ0!zr7Du+^@$Ro*JSve<8Ry#_>rpt*c
zp^opIA(3>w#c>;5!op9)Ys(xn<@j~=cUHHIV7;C2W1-eN|FI3>4S;U_+8GqnOYa|e
za+d9<eXd~UZ@Ii|RvEUI1(nF7d1nzvif3Y)omb|%8)w0A!V0gmfiZDT*?q9l5_-|P
zP(ZSc6&c5kn^<~h`48bFlNRs3@y5c;)+1g$ZVb%%zFwJeS@yOhnP5ecx&O<IZGafx
z_%aeQK4R97b6AtWd5`{F>aF5(O1uj+-`Pu%P;<JDVt%=;hj=inTf}h!T4<x$_wNEA
zcwHn(OXqZ0j_Pk8{TP;aiOTg1+`ZW*IKBUx7dLF$N@aW~5&w*JIqignTCas?A?naw
z8gDqC@Is=v>IFXZ+8$p>tsde=={)1yQA4C_Nbq`&(;1AuW+=UIA69mFUIzP;uZZIw
zYt3WkTG+=`YMos;WUy#8eCk6H<@M~`^f`4dM(oZe4fhgvX0n5L|M_C!nPI^x{raV~
zDLKJTi{|<&7fSlKIsW`Uk`%5=7|m>v$UZmgX2tB75#yi8bX69u?Nc%QeV#|}oG`+A
zH4YU(;_2FN7$_1_KsFo)&s|_yK^x*z_@tZLC~t#Wy3NsBlsz+WIC;7yzdlS6k7#ns
z8Nv)NnAtRS;ZyP;z7&Vb!Eyie|Bj`4#Hxckx`=){qhi*-FU%{YGHz~XNBQ6*zlZj6
z@yljBh%@u3Cj|Q>ljsu>^mq2@3*E%@jFm`p8PtN`pNCR*IkF^tiS>>d2H9z~W?*lF
zTr-h@Gv`0SymonFuA9JC$dJlvS$5nS?t8eus7I;8U_P0H<yg7oNh%LX?dqnvIpy;h
z*L@Mt$89m2!%2(L*t;{a2bR??`gN`?hr_bQf>HL_VOgp(w8QdEy@Q9^Nc+$hgB3uF
z_sjg?h5f4K6_nE!gcN03W(j8Bo^=9t(7a&p5lkANO3!f)P7Vs(=gED)Ik#a!drjR-
zBj*NPBXZK>Cif+-UYNpvj6u^AR!X?=F|y3Mg>5y-OP)#pP>4tG*uK!TcQO%%V06C9
z&*E6|vh!9ra6Zq_mUzfuJQ|a)S<|<C#sh%Zay10Ar(4NOc3y6S-C6T8w-8BvZcBEy
zN!1i0A?6f5iOZRtY5OqTgD=&9>+1B+fpI%#T16}le&lQ5zhPD(^?aLj#8LZ|U+)j+
zbE5q?dj%No<z68kfr1;W=*bSR5!ezaauB)-pE<3Sq;A*gT2_3eu=v9zp{!tjseMf^
zIj(w@v&P^XJVGD|n_gGFrw9URsGp5!1Rf9LmT;C$t6CE4!NwF`+uQ7Q-wqS@^0iHx
zJhJ#y_Q_M0oSF6=XJyGvf2oG$QQRpW3E;_AfU^?iATQ?O_~I;RkRI_r(xghB)JSa;
z%nVm#(W(gKPw!9}=#jaw+jDghPx|i#z^U$EQKl%+m!*Ykl$GY0kXt~N>{Pa2sQ$E<
z{)LUPN>3`)2M;};Yni{QjN=jiuPHg~VHL?A<Rr%G*%_)P?8lzC7ntL)lT{yi+2P&{
zMP~EQVg$bR)7fZISD<u~#>pacT$demCO@8zaMrT3UJqQ22w|BzqEh2jaVcjt#jAau
z@Hq%4rWZ6)n!5k>(^}gwJsh`Jk+{1ew^!dD$z?+H^<OZnIs3s20aRD62$Av%7Y3CX
zQ{#3(s_>@!iq3x$WY1j$i|vPu$hC_uZet2AK!QIPHX4+Mvb*Q*lW+WpwRAQ*IVq)F
zyt=K-fs1n5z2S^Ii@-b(u!6m!SrA1poAMQ(;g}k84o$>df6}ZqzdqcKBJ{rBpjz$5
z&n)C<6=__A7S&Po*j>zCu`G<UH+rhBvqvUz+!8+lv-gyh*P^bymwt$KxIzdU^n;2c
zTM|mh@B!Ke$ns}i0pCL`^v(^0>_$H*=eX4_bzjIK^oFFXWbDti-VGy)#@=_<8E|H2
zRxte76G6HEkb!v^rX6w+^ewprZ${>^pAHoB@>tzT<bplh;iCla(gkMbX~^wFHOMM0
z&eOaip+7gLlgz&;lHRB-ZvGa*A2pXHtkN0I&A$HY%HC!3hx=KI?rPK_mAe6Vou`2B
z$E1FU$Kw1TUbT1pn+_>wUMlCrLlIImZ2%VsxzH7|T6Suo+p#}}+*@G%EP3)aYuh&$
zFdxwChRv^E&AU11wx^-^CpI5}?_1U)eURH+634nK7la&VqUj=7MMCmgEKBx2et`o|
zPoN88`Ns(qqz9O{Er@`>s*DqP|0?kR?8e1B9gcoKq^Q59>>sS0{+E&cS2^a-<o;*-
z-@j$^znyhzxA9Mt<-cX~pE=E6RrYTI|Icpxzuj3o72B~7M^*;N!3RO`7FH$Md@a0Z
z43a!jgnFY>KbEM&HTZt$X5NK>xcZ9d5AYVVR3`cusBqih?;dtzUU4M2$Uektlg6jV
z82iGT-6wjwi1y|1%?#@H4A*0EVw`>SZDjG<DN!Ry<|-UfD^U@9w8+j{-1JUc?InTc
zHfU_{)bg#o##LZV`ez}M;(&Y8z^az$<Pqszu3_tdSrh$zvwFTUp=<k%VU6EG>I_CN
zfey!KbC7VaF74Vo2L}I+CzTKWF^2Rr@Mc_Ad-Gfv<?xPsudCHuB!4<^P5|WRfH~B>
z`SqrbK%+T$raAjZ<mfnRlSL~37|x#=Dk`>Zs44Z1<acHU9p4`z{#ttB$&_Mx8GiC4
znZR(FY2pSQcOh2Y99}ieOJJM~x<UU*oBovA{_YY+KyQt<-n2d~T{n>1mUpNhm_jLo
z*AvPL)Tahcks?XgXmVjU|E~UY^Y2PcH}{{mBi#Hk@Y^-ROn&7rDEnVaWv2C=Mmk3k
z4$|+MP9y#9Cos|@nc~@=BYv@wcRm};z&|3@H(9Qk>i3(5+>FLg$Z_>0jt5KCXO(8U
zFYMUS7VUIV;*X2_XH?h6;L{_!o<F%TtNjGMQ^HJdZCKC@0i{&qzIKiE9S^6?onL9R
znQV9p-5@$`p=sQ624Pd!Hs3sR6kzi|XM8qeb5vndG6y2ZI<unLe3oDE(}5pLfo~sq
z<2JH3%prkVP{aFWBWa8(R?^4bP|qL^^wdE&)RgD=_RD47_28NzK>l1tB-tyEbzLDi
zDrhG(TKy@efXGW+KfyQC)u_r{55w)xTo#<Ma1&kb=^ygv>m7PKGWNf34vdS8r~A48
zX1syq6VvQ@VCmSN4AcxZAyt|YwVkkn$71-Raup@J?Y#1MMoS4+MMS-O#l>IQ0X#K;
zi!P2@Wl7ArHh>vRw(B$Sc=-zo1UPhG2tfiM=jXA%lCtzv3jH~?G?WAy4DG@_6`s`R
zn@?|;4^O&)9g*PU4?AQ4^tL1JMsG{-DcE;7=lfSoU$z@X0DEP0glSP=x^`U(m*x1A
zCLEW57|JlVGha6@WQDT(Cy6X@<`F!kWOX}}9-Ww&nAaAk&O{vC#rgj7Kl$=^!_fwp
zpM0_TAATn~VKJa2=;-jZthL!zQK>}Yd&wWb<x;THFH}Kau(|6x-Lcu))BODP=2mp+
z`|;G{o^>e?$iWXAQ{G<Lx+%l%K3>hPwGHlT7c0H*cAR)?rEe(x|5kT7&;TcuzvTJQ
zwZ_!_++K}!YMMICk!+{Tv5-99PYt7=gal2U>$<~f?d3h$7WO+wCnCXBiUfYQU{0ni
zM%z`xLQ3}J;$)j=_Z!}IS&uP<$t!t#k1VS3M;1G%n@p@5=qT|)Oe(M2lA^sOcaey>
zT3TD>_@{<v`%<*!!AiUo?_#-Tnh*`Mw>NnJlT+nw@F2VA!8+NL{?uyi%nloKVyMTN
zaUY|M4QT9#T#J5YQ~~s$J7qGI?4iaqyW?>`V5O<&v-?dP*^0qXFWLRKL1Tc%=8i8v
z!Vqs9>R{YDnkokIoUk8TJ*m8UOG@|NO*z*p_gQ-?9Zp{Q!kj2Xx-}5z<D>Az!Jx<&
z3f$!W&cgu)ank12N=1qtUs}Exe(g1+p+oh!1pJI(x`9ZJ8`)V|_o%gf^@9`FdE!EI
z6gQ)ROR#SZzmjTPq*BmXZ;$(TSDIdWcE9mmS8Kda_nk=)P)@m|7&K;*z#p{Ps|NGC
zeWT;W;@zq&6n02hAB{y##;-+3h_mIr&6mYJ4Pf8Y3dq(ucN5vp7Qv;sxiPBL2&}L2
z-e?5s;NBOC8M_GJF*vs`eBCy;Ujp$}-nLUUiB6uz2{<m6E~4s;a2AVOFrGiiS;$b0
zZN*Sp%Sgk1t2{k~3H(T%Hou&;l<a^O$&StTmQ0@lh0E)8xGb(C4PQY^uXX4A8oso4
z8Ei5-J$=<F+OR(2c@>nZ0+*ZfJZFPV7)le!pc8K)173~&p-}C8nj&_JEyt?WNnO-E
zYQo%NixBB{u8()0dynhrahNhU8?+)N_7W90O3zdfBKgIr34cxTtn${h;+#s6U|qw8
z(nPl!Ttc9vTl{^Kg-=RJ%5O<2!mYjx*GUY8W(13&NM2^z;&_9cG@oP3s=d8G^4c~3
zt`CzftDX5RlCll}68p*aEAV3j((?aDJ_r=G*F~1W0ot6W7VU4QO&cjsq}~qa%a}|r
zZ$lgx0#{r#)tNRKWaF&%Zw;OkJ|g65$E?0dEoomQc0;h&b^0T1R+S&p#NE0Qk=Zvy
z+KL8`+&z><A8J8cSwtywU*d{!&q0V!o@6%aAsL{HkiP4a`v^1YQVTHqSo|pV<Lp-l
zPbf1^(xsYJQ`G5rH2Wc#B`0IRhfHLxX`O;H^JGI|bWJm)N!j9Yj@r4Uux1>7mWrG?
zFy8D(_0Mdm7^l)a66s!u?LuJ3+bWivTOv?A>@P5JSEB)b9%{!$=vV^K!94UhHU2w|
zxFup0iexA&H=gc)R_csURwsiT-hYgn(W2Or-C3B-V=;pA#cu_Ngusah;R88Ro_6=L
zn(i?Q2DELx$N1aERuHs4U0`_Q%>ALE&eIb^H1pe;V?hILQlI<Gz!Mm`HGNmSCRK!=
z`{=6X;6h6`hR?fUEDb5uw*vrVG5h}lY-_-O>RV?bIWA6?Qp@kgvkD<D`AERFMNV$f
zR)3on@$j=@1Bh^q2BpJPNmkORTpJm$lArr{J%CNJ#ue%;8H}xuL5>f$(e`r}AX6UJ
zc@D?<;3bx$vS0D*$%9>vBj@5%;WP+_L~b3+*t@Gql|}F4OPmx)vRl)1UCMa}LOQnS
z*o3(MV672|G(PLk^%OLSqe%MZFKN14AZlwQX|b2MlP80oSvrQ;OL0qL?FDGyX#{Sx
zoM1LS5b1(%3*T+))~T0pLQ*~GBh{TRPSkRPOjve8N{NlN`#BtxtAqI<0;x~xKL%nd
z%7hzX{>YH@kBi{Pjj_Krd$|tyGYDnGcr}0eA%v;6K8?HogbTqT6@18JF*GSvu2$<J
zypdS>vNV^z4AyIBjjX2F+K%S!_)lMtU&8e@IeB&1i1Blo=^Pqw9c>e#E|N@NwUt-~
zcEK3G^+cS$w56y!8-t0-JXZA@2$QSbMcbQ<c7(zYKAH$*0<08@WFZ_^sjl7(X-pQk
zTz}MYlX;%53`HfTT@WE3(cvBLXrhKQ1~(4SJEDODD+hkgrIA8Zbt{6(1zq)m^hRpM
zA+l4D+!C5I<Ft!YFJnt|8j*(4w9W7<u4s$u<5o_|nM7o9{r{XvCLn>$A~}X*y1Ur_
P@55mC{#~&<EUx?)Nj71{
diff --git a/public/07-basic_statistics_files/figure-html/kd_test-1.png b/public/07-basic_statistics_files/figure-html/kd_test-1.png
index e5054c2623ae5d41a96fb736da8f32296819db7d..98541413f2326e07fa4c167ae9ab14a5b05267ee 100644
GIT binary patch
literal 16007
zcmd_Rc{r5s`!IaZ3`5D5tt=S|MM&AQP9l{=5g}A$-?Oh%DzZ*xDZ&)lm$HoPm2BB2
z4B4V&8#^<YvAy@r=ll6C&vX2q<M+pV9MAjy#>_o)?dNsw=XqV8>R;7n-p{)q008Do
z7d5W~038MZ5T6(kV2%IN<p=<PG3j41)B=kDpacLmfRYlR1U|8VjSXM}KJj2JSOrx>
zFB>H#tdfn5k`4I8D`Bxp*g+d;7+3{>ZD50q5*BKQUf>VB4Hj#I1)qa9c)SfBeC}YC
z04z9*4S?P43N}O2?0#Xfcr11hi^pT};ByB%IEWqG!P@|M8znpzeA|G{U>h{k?iU_E
z2>#&*!RHQsXQ#Wndl0KMh_xBSgKsR<0=9v3?0yXn?ts0($KcM+4gi4j?%=UIgWwy#
zTL(@6wY+ywDgu$Q&+VegeE<*$gFdj>@uhnJfC4UQo-_1)JvZ!VHP~G9(w&f?*Kkt}
zf0XD@?;K~Uwj_TxC&fzKp_2Se;Z}?4bAvD<{pNOB#-UpbpB7Iemeoh6s5K_Z2lv56
zQyaVvwZlHWXkOz37Jb=Kz)!aS$sZiiCG6TnY2ehaYzj7|b;M+gHY~{6)00|Dvr~9Q
z4q{O&SZgi0ct&ykSLFE2$(Let6$X@V8#6V@jghA9Ps`RfEBvWEJ1bSIqff*tI@)6A
zcnC2^(RuN(T(!gFl_YLlp2GJx!$rn-d>zjStpO(j?mg`FO?kZ1BZ4MIy2zo&>lV+z
zk3|6mM!_m+o3~VC=MUnl)>=6FSoH&qI={>;<fR!*apR7QJCrYbCA7T=z>s-{Z7|>F
z`e^}2(QrnB!#&KmA;YwriNL$G;xYthoukpOei}WA7k78tsJ7#|p9d>;`|aQXeYvr+
zv#8w>H!v(_u}<pf{>I+XKZAk6A!<%VqB{Ky$49<ZU1E*M-x;Dz<ebOg?M>V4Zw0&r
z_My;uH!4oxRWsWq=_HJ1i23rrs*Xw?baZ1ORKIMI>w7ZHsvl7Ge7EC081Id=M$e)t
z^f5GR^l$5hdvo!{I+)_MiisEd2>u2_Xl6RQGN*StbMJ5Q^@y#s2=c-zgNwXP$sFS)
zfm;>9X)0ceBpu4_pp_F_ex5k*VpE&;wrEQH__j*>=xD>G535&7?6;p+CuX&TwtHo7
zi6(b<!7~~Px0Q_RF0FO4L?qg`b^l(|&kr?jY6-E7bJbu)dpFONQr_lGe*O4yFiEc`
z%fOPt!;oDw)YM<pIL*3zR%db<x2@NmN?le+tKG5ycOwoHM(5p=_&w+|Zu(ZOq;vba
zPQ9~NW?<3kceQGX5RaQ`*$h*KO3hA+b#gA6chvZ*H@mgA1M1>7o_t5u-4ZpUc1H9S
z6`jvi$gxKrcRZ)mN}DkZ$Y^N|JPkpIEeIVy*uRlcW1j~U)O<ACNhlCgqHBIXR03?h
zP3e#xX-22DuC{gNjSOwPWoSMX53u8i#d7nWw+d59dSTsniTb&r8UxzuTXuE)y@IaD
zi<T*aRj~deKOQ=JbZd4u71x}n*f`E$mvkMHU|poFnvN?W_5D_vz~Pbev==q$v=RY$
z>Gp4nn&w%Ab&@|XE|m1$cIR5nv%qIkFa5SSPWf8YMBdmjeE`ya9F{bmr?L83g^1`;
zc@Sh5D7$qmsW(r{@ze5owK7jG!TVcQ0XHqf+FY-9@7MS9z8H|w$al5s&VKBYYL9U)
zEg8><ZTKzUc4zEqFYL6`ntSP^)tx7+X9g2ZW~E<W#(+%hhw0jxoeT+CF7fshH9NuJ
zV7zsDYxqH}?OS<D(G#L;Urd1wga3<~olTGW#&dJMLMUCDq1Oiw{Js&>1|niWsdQ9V
zUd(84SR%`!=IW}>PQvE2MMx7(MKYL`s@IcT8*EUJ+z7PuSTI;ue3XdfelUE_Zp@<-
zL8c9=Z0o2k1n@Li7bjZCuS{1-c{i@{ToplsY+9-ksyChcrnl3c_Nr?DmnQX1?a#FK
zRrS!qL`}0Ia_7hClP%8yd%bohliG`3vX?gA(YqF({cr=8<Z$hf`qr=W!GDI>Q1t;X
z9dwRK=qjCU1x%#ozrEGE&4pU=5H!|m*~zadt#u!G?dL4R**RpXxJ&0{;GSS?_w3R8
zKW-_vswSCGPxSS{w|q(|4w+hFTk(yVtHqMI^_f_Pje$r<n7DD~M4`b3xP81423c_a
z#~g6@XDD9JIDpj~R%|}>?s4QN7pCTqLzEZv1u6F>(FL!H#zi(-x4-1jg@byB?znnq
z@;1eAqq@&{$cb4zR%$Adr3{f#ba%Q)R&AjbZfH+GRl3UI+?NB&8iyGd?)Q(c-GXW}
zj!w$xCu{na-FZIFJgC8jn%mo@mhLqy*@tnPBBceX%Oxdl6OOLp0eTxuSqD+@Qqv`&
z%R9+)0U~KH9p;<@gIf)9Xtu5VxK&tlBZG4^(9jnWXt8B97gACn$y1-~bk*qwhFqR+
z?$o^>JFVKYb~;b%F0_!j=|{%ba6`*^+H%jx#WT=`^r|SG*VA@csUht6h~{D2w3in!
zxq7<PzJdMDL$fk;Q}0!$8zyfrbj0Sp5V?WL4VN`Ct55F9_u_WiUc3-e)etUF0KEEn
z-YR$eK0n<0!$+R>`xN6Dpe!*!0!?j|+^<hX&U6=UUENa+@q%y3V@osXvoA_)FcF;z
zW|)CK*l_Lo%|KJ>8C!=u;KocB65dnIVs|8Mrn_CxiN2Y9gbr?fb0~76|13Ilyp{^Q
z%G|{YRVZm%FpM<yt1IZC9K6&Gk>wt9?44-B9Q98kui)k)bs^dc&4jQnXTpzOo~o$4
zH0oDzd^NeGTpZn?*H(3##?BgM{yrW?s%5gdH9v#QmRl+FDX58Z-_>F71cl4)Ch+{q
zsWXI~78p7yXvvEsmYd15OgU(Mi>b)WN>5^VqbsrVRw2h^!{U8U`;JmfuGmew?3WLz
zC^8A?mzt+v$&!TU0k0A_Dq#-on-4<?6ML+K%pS73o>nN1*6C!DW+gm~4SwgR5Xhae
zMa#?QK!tTL4t%zrLy;-er6M(ZCeaQ0NPyUJw)}bV=))W@y?SXTg3Z<X<h^MFC8de>
zzAJ`sO^lHQ!q8$YK+)CmaYD%TvUGVzx{K<~Dty%{6?O&mlt>4cRwiFcjlx`djqkYS
zqNrRnV19up9baS+U{iKGAD%Jt5NEF8YpUuHAoRPd!Q7;j(FtC9yP`+{H_<ku+)r&=
z_)F~&{f+qPa_5CrB}49fu7O3u$+NASG&A?PnVE%^^~5*ZGx4(uT|Awx!57u1`n3(2
zuZ4sd)#Qjs>^xM+$w_Id4HlF$?jv{h5Bg=sXxy_Jx8>-bw7-EV-e4)MVqY#_{ZOra
zj#YmfG(FZ{7;Z`U22hr;F#usPpK8a2>)qXM9z0Hv?y|Z*O(eg4<3L&7ZDvPJ>R?zw
z01I<kiI$Z{`Mk5rNkO^7O{Sk2nO+LE6rD)d3(isxdY`7eQ%EN%h@K4jPP|{NgpTC?
zTnv=`Eb^-;62uYD)YUCoDnM#X98C=q+4|E{_$Dn$?PO$c_-Q3l5#_t&6~U5;^1D4T
zy*4Zn8`g>R&6}cfPXGbDW!AH{^UjCILoC^G*C>DPhUg_Pk17!F5-mL`uk+b)^pT&$
z(G%mfE|>#niY#wZiha-R5w4rX%diQ<*Mux9nitnjGW~dgQ|ppt7VkO7Up~RF>Ui?`
z%^Vl;mC;Tnm?+3tV5tG0=YJ;ujF#7vfP0!+{JLHfOd(|D;BS`kn5&A-&^1>x=EJQm
zK(20<U2(5Kv$nGHv>EJUgPH0w={p{XaZsO_Z*t1+j9h-!;HW82H`%(G(BXt7ob!5o
zXOI#*ecPC!?1SN$4h6?c=?T;wsSY}nz9lMG1~6Mpo&VZvnYEtUJD^-R0n&_wNVJ*n
zOr4BGvtm~vf)xOVc^;T!qFRIIfLAub!6_X!#h>zk!;mhc`^5J@<da;$aJ-hP^y-S*
zNZ`ooBVhLR9CC^lWa#JhfoKo=adz+kVK_P4Wucop3we9eNd{f=CZgl>@s(R!_uMLL
zTA&_Gg!*d-U&jt0j4xqRH>=C0_bN-{`QkI`3v^jA=^>l-eR1X7xOH<*IiHU<*T_*N
zchDnrhj7;Sv&!o3`5(Zz(lKHbV94YB^n9|Nl_g9_@-+-BDb}p)TG->&a(PuCM&kfs
z?$ika&)kEm4E2I&Vc_TELnyB!Y9~6T>NTk+st!IYSw!#99mY||-@8jRj?RGqf4cS(
zmee%q{}3t0k3rIo69AFlGc8m*S*2IE^aXKQUzpi2KaZk;>V1mnsBlKIc{c2>rsi#i
zM~6`tH0iuGu)9TtM~8Ncn&G=eZMwap76w)<h&|~r*qe;N38C*>VS0UiGHso;%g144
zV+?HIYj}8`v46!T8|sn7C&2lx=2tH`SwV=7@z;JCG~XfiX8`OVxGUrZ&0}cHBNU=1
zNK+F|caiZ7Iz@Z83XPCmP!tx1kw~PPft4UyZl%bCx7`-;;*xTnnO){IosD^+zvZ&1
zU`(6s^%n~ELFvDUA50Z#J3NE28JQGGYW=&q*VWXNs&Sbiz)J&D`F-_4Yt2Xj!<6^S
z?P5<%#B`xvN?dzFyZy|w#)^Xmm;pWM5fseSvdCBh-lG+0WahX`^d5wX<2An`;tao#
zgiB7NuCYh09wq9O!%G(vn+j|fFpCIJRqAFt?Gjx!Q<>-ii;-g?m3pxa_wpw32t#YM
z8oEW#DO0gM_VRX9^4;&*ZPm2r=F-J(PqL_GPIFndHgkb<nA?cB=&wnFQ9n06lKS`<
z^%)$z{LS+ZPxtCq`n#M(kHgrgfXIEP4;)73Ice2*X=2XrpDcY*u)Nha66{Z1WyD4o
z#Dr%ft#72GVj`W1ecsY&%#d7@+N%j|c|1SP{TTx*tayN~OH`&k;?3~!(#6b%V)?nO
zte&?t1g!Xd1CiwYQjlO^_D=dj`kk+j!#y&imG4As-0qT_oWH3a#9?_PBkMl*+)-m#
zL^p@Jv|566VB+?cj>+Rc#Fdb{2AKIy;v=MVvE(oErKRzez;4CiE5S(u5(?EV<Ph7?
zzL7kDXepUI=`U|XIj;~YRWE4U@X}$yw|X>DH~E5<q@wr*1Gx2_s-wae7)WCj(TzUC
zD%enjP=+eML9u);(V$>(e22G7a42E&8~5hIn*;i1o7XIYBX!-~R{ONa@5KGK{K1AJ
z>ATQc#YJ}3%w@)Zm#wXo_xQeOc#*cXby4}QfpT@>WKkfiBVG27p;z)6ch_%p5BJb|
zuhWnq^5$Z~*F|=GjcpI-%RgYmmRNr%Z%Q|GyuBP6qSV#b9G33DC0C@Vg()3rvoN##
z+?x4E!cG}Y%gLzUOboHCei0&}C!W&5>{{%<tUmt2u67{s=TB8LG4%A4K6|yN9^35<
z4-OnvG1Th}PQ8)_^#9ln4<D%gB>!aW2Nl?=Cw3V-OOrMQk8KG)9C&z*XJ1Zx^q&!1
zy`VeKmDa4b`%b`RJFaeK%)Iu`+<$%Pa=$^DF4hRs<637F;MYg>OE0x`=<>05?2#~M
zZr%DyOLo%Dr8wkh1zcr%r!BVPm|;V_$;^agO_kTaPe`-4nhF7cu^#wUIgWT<ruqap
zyJmY2lN9D&vXhq;%E{T}h)gRze~2koZc@EDq|SfBFyNg>Nyg1JQ&{fLYqegPz01pS
zK@>4cSnu9aCuY3VC7Zj;r*-nyR`7uW=P{(osw6#W+gIR9!I<ajY$*4F9TU)%?|nP_
zz}~7S4zGe9(6Q;vkE!JjBH+;pujyOWkwlStVp7UZHdVANC~e|1WgzU^YLD5bG(!_n
z<hr<+L!P*5fTQmlqn;VP5JiU@zP67J<7Usi*p%41zx~u+@a4@I;H(**K4Xr3VUpo8
zQhKM<cSMmD>x~y1&L#G4Zfa4Sv$Z5Y?$E5ZDOZRqW+4u>Evp%(MMgE!W~9ZnV6~+$
z&5QH8R)Hiv!O4YC5Oz{{_>*o_RQU0Fw~x-Z_z~jU%**QB#`etIDPf(JZ;L|g=}XVo
zQU#6aj@K$lUHw?VqCcmL7bE)11;vy;e!q+&-`WVd=Pv59*c;dBPczk%t~!r#a&j$8
zD%);YaH`{p0wMP1zG7D1?ew}xk<}8J9oSF`AJ|`EgmEeRU;?@ZNrp2-@>%ix!#L7@
z962}RTrr<bM6>7j<x3&@HTU4wRsnf8{L6o;8w-Y!@4Kv(-KJ~q@0!6ZYv-z4)L0tf
z+MQtcloj<XHy3{`Xb+cTOpAJ7rvFyW=fZiPWX|g5o_)Zj=Q>nWX*Kfc@LkLVtE=$V
z&X;LpxxtiUBIx;qTl>t_k#SW!(?O-9A#8J?pU%>Dl9^Ci9lUfW1t?HN_V}0m4qI))
zw0&?A$iG#}Z~>##Kj9GXw9<npE#Gv~VS;&N`4<&M4b<FjP`_--Dajjo^}3WIQ>L?Z
zfjjq=K}>UbLjnUx^O+Y8bxH1Tb$VQ@n`2`RcRTq${54J296jW~wL?6uPM7_n!KZ#w
z!Z;XPsATe97#&%J5HFfC@et6DKj3xT%?m>UBh#GKJg$}M`CmM9S9W9r9_M=aWXd#<
z0`ARaB`PAV)BEzcU?gDy)~T<jj_7mN(2UGG^I78J`Z)P~ZeKCh;Ay%OF_UO)xlgqt
z_^k4u0;8W?u-uv!&+9w`Ae$uUS<R&c#wS0hd~~XYBo>s{J)1kR?qPS6ExJ{fSx=e`
z#aCldz^E@4c|r4}Yst6XY<G%sgQO*e_=95cK6~{`8G{i+_L69;?c`bw-on`;ZPUSL
zK-F!dp?hW}OFCl0u{_NJg_ifPF=fZO6#g`v$yO8f4BoG-e$1Fzsxq<A35<~f6arPz
z_gERk`fEEPN^VmWOrl)Pe=?5*mIWqphgp29II(7E28J^cy)sj)-IL=QLTJYsih~xW
zvQj=o;)$tW=^O9Dk-W01Lld^{m~Ts%Z-V1vuz1@vhcTEO+%PCo(>uzCXusZ=T&?pn
zL+yzE7%|S>S=zpAR>a~)ljzvWWEnM&A-lq?*WdpM&_4xD?X6kn9V6-eP`=$m-_DQ`
z%MK%zk-p$aSCTrV8;W^3c$MJRnizlG6fT1Ri8b42!J<QbqHKKV`Q5H^Ti*E|M7`6a
z_8iTYG;cxJX<_u1w|!+Kk;Zpf;g7eSUplKeU2`}<84B|6&0Y%PHOSidQ0YN+@#Taj
zfFEN+9R<Ax6#4=kBNmN`Vq`--01mT*sX%Bd3>cGYVt{?n46uX$IiNYRaj_{GcYz}k
zS@5B%P2EBDV4C2eu<b9I(4>DFCJ=^0WlHPAykN`q;T4*H`oew@Xu|3rzL?TqCdmp>
z_>dldO~D~zlXGD#3|i5`NeSK1w|1+qPP}h3z4vH?!g5CJso#y>dsAj241@D<wC(8L
zH93Xg9{&v9`?c%d_D=kJV-?W@ea98*1~a$PSC0-DcS&#idwZS-cZDsPjdYHAk&@6i
z!mFNy@0;2n%XXu7)RI)tjc#8P83q*A<4BfY3`PvPYP<9TTIkhHms=AoRT!C|t)K0#
z9Vqe}(%g8GSFP1u%s)-5IffoTMF^l6{BDKIc6+~Wq|_7U1SP5W;H9d&BXlk=rTu*6
zp}&#(c~;LrQXp*Q9<1xho1%pbkjec<Z3p)2k7?@!S6GNS9fV--=rYW1=k#LNwJMM_
zo^>+2m|}Bzt-HrQb5k+uEP8rhPOO7^{?^;bX#x8{KaUM)=vI8e4^9Z3uXBL-0NrRJ
zeu%+wo^njRxz&n=kngLT$L%DJUV6!la~1;0+x#ja#og+>M!d+}MN0f^#pZ*@FkWM#
zy-Ui;4{v~+?Zx&iGYtq;XXc{3Hkw;xnJxv<*X_{8PMPwewhq1bJ{Az)PIu9q?kmd=
z{4O=FFlo}Xz1A~T)LyH+?Va@q6|Ge|;?R_kNOv(S4QuMn=6@7cId*!=Sr}U0L>1jx
z+Xs)E3ytTKr5T@;X2Na<aD#rX$tka%fY_~~f?X*4yCiW=g84W5N(5Ddmg>Xmq}7q7
ze>DC`S~6rvar(A%p$GQ}`mPDnqo>!Z`rh8p<X1!}p&+z(W;$4eY*M-{<oXd}=w_4P
z3+DFb`mYNIHyUyawS+FkbED=mjJ?CMl~`sMw`K$u=ykF>ex=Rr0@&~DmxD`q`%Tt=
zxo#!NFmH4C+8H!eCtEO^`<RID@;l!Ll=4fJ!^~+m&pxn~M?z9!I}IB%w&XT1eZ8%?
z(4{y?AK4j3lp&F|sm~J0BHtw5Fu1-@A4=<YisL*=JITvB^;Yf(K`TfFg%80;?@n;(
zPAVy@m#=3+>-#&Cl|^@U9I138q|>L5Y+0swyqovW`-pEOc6N#~@dwY~a}0uWBrxTQ
zXqxM-jyr59TT+j4tvH{arCy$#nrdTm;gDGWxzuGBqDkue`enjmz6yt{m;QQU<EhR8
zBi$A4cJ1H46HMWyT5rH0t!A~~&+E>HC;ZPz)wST{*+!M1s;NohlkY_L<<1rIpMc;m
z`IY9|+fLcYo$npG<CgmFUHsN|j~jW+r>bs9)N-Y}Q;xz0B2FPP3M~ryOE$?MK-}e*
zfpf^?gSXTV-Am2d*(p|z)WN)FN8MzUK~JA}0Y!P#nwXiR=nh7<laCJJ>?IV$FBvsS
z?Zcjd(=ox2&7S?UCry3NJY$D>T)^BG`UJdk;V-Uv1<K%2^!q*i1Xh;S5uci$!ll6v
zU@IH4ukp<k^@@N&5E~*cSUNPAXC;M_T?^_nw!4_(Dn^EMR*vzk*~^+$ciw^OO{f`o
zhoJVl{*f6;EM)^VAm|7MhQB(kE{6B{%7H>_;9!@(_Ptp7Zs&Qr!tGFOD5&|qFzbu`
ztsjNRK!c@Nnq=7f>(F#xf#DI(mWppkQNEiYW&GfTU}U;nTeP09#A-t=xDLVLfoaTD
zk=8ycH~)ToZ2;7+8lPa-E!(kk0Xb$ehuGPt(l7|wLKn1ih|I)(Xn+gG0de#+KLbJg
z7cyHO*}n{H5e~fp9*>}X=<Kdy0Lw4Tguwi0nBj&}haB53(AURb*)yb`MbeSMu=Tp7
z@b+NjNMbTNQ2G!-=QHy7i+P`hy2bWvv^4!8lpH64y7Yd(KJ7-k(RmGU)0$Y=<#$}-
z@Y1z=;|qPzgdbtKUqi~;SYdE$70}5={f8}pGwTNi2QDt_5#NQix;EGy1<`*BL2iA&
z6m*-8aVi~!kd>6yJUL&7LZbyAAHoe<Z`dx3wYR^bxh4p}K^VhfN!(^?KCTgPb`&la
zB&0m&Tbk8?Fw};Dby=nFFI5*qqt)q;;O?SWpxP<eb?pp>G)6Xp4h)&S>UC)PcDP<G
z+ump;CcNd&%!?3k2rCrX3lQ>Li6c$KAJ#LV8G|^(Vqd}9L=*t4;k3p~Bu2abswfrU
zOoE7efS^)Kd$TI3veb_S+d1qomH#Kx_usaa|KAz_`)5|i$5#rH5nkNP+O=`w<GJpA
zAkVh}!!=Jyt7`o<0L1TyasB#Qlegpi3b#)ymif~|X$ToKtM(-~H3!yPBIo><OWJiH
zrUHqM_eg*4X}59^;_};=-#*uRhJGb!eLs_2a%ZpVd3#FlcEa?Bdw;tk8W`AR@6X8g
zlKI((uprLcn29kX=bk%nf|3(KwghSGseENWSu0>vYFf`R-R{KR@+6hO>#x6uJzX~N
z7I{lF@rBccGoD7TwB8vriu!C;#RCmJg^-Qyq|h;6by^4)1IuVcTzHY#Wcv3hYmZ;2
zjxazw=fhq7CnTSs91d<lgPHB9FkBhc2FM5tzvp;(04i32ssJh;)z~d63xC=zN`Sn*
zg|UA2IX4I7Rb{v=xR5jq8P3lBz?}Uy29*9ETtd4=0rtDQMRrhwma?-qe3Wv9Qc(My
zz{JiYeR^J;pZ-zd2;Js;XAuJb1{UP2J;euu*CzzHp}PRPJK;<d6s+Df@LIH`794Vh
zaBG7JJ}lo6%#JG){ev%s*`7n%nWh>`xwN3rj=Ih6euzR1P@i&H{Zmd$dmr6OR2)72
z70`{86qJ(>G6D~3dLQxgD2|GM0_=|p=c;E)3PYovA0RrwjUGw9%g+LJd7YpzDzyl3
z)qm%JmMg7>`2_9P-}E~weJi$oPM?`x6Vsc7ZT*c@nlivVzg+!(``|dfvHWh^+o=|w
z-8B;A30j?S&WgZi0o|o)+qtx)_n>soTxkwAL8OpRpjYx{C59P_YD+_G&Y3fiNoJHa
zcUmr;9pw=}H#oEio-zN+KXpWNshvC~r^{CZ8ju;?<~t?`bMi^_O71zmf-sr<TP!m*
z6{TAqAa`7xvcw!*qhfLc1kL$4Ew}Uo90KK!q4UJ7YhFKa@js07>ql*Tf2{!71MoyN
zZ>?c#)_Y{K>;#BfLK9#vFsXdIZy|^a@|68~aDjtjBWuOmSI3!!Kyn1rPbrP8<Vw+1
z3Jj#w4zNMX8DNSf-n=j`qFK>hbkaB&$xR?jqT9PITC6%%UNb@pHJHBHjkPiQYk)m8
zhZ=ph_CGms0nWLT=6PUpXIV@=5j_buys?E<DtBD<SLN3V9obhWBmi<RSpr@1yO2VT
z+I&Mc%Y{S`#LI<)eAQRiYUVowhmi@8cC?_k<G1J8irN}_^lWdX<H1?3<YUR_)uhg~
zL8H>(hIMmax^)j71|i=I76@4Yv5w%#{edaL=iFcLpFooaPkswUlcHYMtt>)_$phXQ
z?^M%#@84g!!F=G4D*{B1us;{hvqWrJnb-3Ph%6cxj%t4PP!RdsWIJd(u!ItzeR8Q!
zP&@rQYsUy9{xDQm1yB&+e*QIDA(R8}v@!eD-GqLRFtTa%76HVo_1Z%b&;ABS938#Q
zhO1_}%klhXDJIhxmL&S!Z|PSiS$Erf9Kx6$Y?-39h=dl%0&<>Bss0Rp10g<@zIo(Y
zn$jMXo|~V7TazCCcyx?g5;tvt={Yymy_&AJhs>U|pi@w{8<?KQ=+GB@OXXb`z<>$l
zbtRNAk1)OpB`@?;i^*;;b|Q<Tn;5*L840@pX8{yONPY6-)+c7HPg4WOPG|0_97x#?
ze8@27!RtX<X`6mZt>1yF7OOnGFJvv#T`{0@`3UGQU7Rw*NTu1@)*XpityQcepA;~j
zLJ1l#`O!qIyc2IizDy_F)97{Zf`@IjQi}R|wyK+4)JeT*sHG0%gJXjmeRmEr$;o49
zjL(P_bu~v{eoo|Rhnhh)d14)u+@mpmo~wDT!?62|Z2~AkyS(x`Cj2G+MY(!sU(IgG
z0(^PJ!CelCbE?c9Q}e?-Zx<>Dk@81HsEVH1vo{MEk8ei&QIXgF`GME_Gbr`L50Oet
zr_c_1g`ns5YaTQxARz__QaAliZoCTS`flDOSoMPT7g2N*>rbN%))uB-WhNepQM(eC
zRxH6@(D4OwT8iMZk;RT|H^3xW6)Rg%|9S+`Zn1Z{9dxkJ;09QT2h?(bhLER2(wc@v
zo`qs|Zg)2>L-T?O)h5l^b&<zQls|#8yeU_t_>~~CZSB{uWHuc7G#`A%!3XLHv7*lc
zQD*GH7MmKM7TBeRiSPt+{{)_J?>QEpAqK}L#ra-3EhdE$Kurj#C&&XlNCG`ToU`AB
zG>FGJaqCUjhFOPQE(qi;SMx{59G!8D*ki!~0uKi_?0Vv9n+*$+(SZw^x2!azBKf9J
zqd;_Cz}cSuAh1eRD6xUr6XeqN1=s!RHL*xn*P+_OM_KiqZ)-joyXmNLwUy7UFAtb@
z341oaW%k&^kky&Vf%@afUv{VJC(=&MtA-~OIIGue?}wPD@!_mFvIk!htO{<oM9LhS
zA?>R;>4FR0MmGCtYEm?Bq`vYVePspd_IvIK0kQ%b&nNkA)_S0O#A35{i0*g5=icN!
z1@4N;EJVPkZCAP<?Yfwtn%x~Di(ZQ6eb(l6v3H%e>3fs@_*Sk>6=GnIuyWFxv+}|w
zrf-DN_)sRuC5}Z{E61#8ELj#b$?tL6)H-hY_t!b4sdmU%|C|2%xAX*LS-kEp_it_N
z8y*2Ya5~UD6u??459stR?J@&&oYK0d$Kqxm<`j8`Q16_kS}l(l$5eqiKR*p`Cf=!h
zJy=qwZGHCo{vvzM8t=gIDh0Y`XN?#pem5e>FpVg_!Ql9-4gLpc+<OvFmN{@Edg#g6
zO9f<x&$&IR4?3VZ$<D5<?gu*QhaZ5x`;k1vo*E%3U&YremRsbtib7Eb2qLM?*d$tc
z@Xl*K7wvTQD)nT@a&X}W?{G4!<*GL%)?ejAn?slZdD*)vE_C*bz>NC=g|NSLCh>i>
z+v2OHZE#<jo$$D0QwpF_v!w32>|`nQ78-<TFL6WT!EKPY*M3a;+yy&d7=e3IRjS;#
zel@RF(J4egt8(K??*54iOOLUP8erbD8{HXd0Xw<1`X<8(pC5<&@q86M_Q4LO3&K+P
zC^*?^GkZfuDK`82KB%`idh8<e#RU3`=3F%++g#uw6{Ju__*6+B^^V}A5YN5-Qi^JD
z{Uf+Ss|fHo>2KgTQQn`qJ^#gopvZR)<4!QViIfW0y?FOKJPEW<-1quWxj@dF2a3#T
zUjOJTjwTJHeb(l}txL1>2q&1cC1bz7rI)SQufJs$U%{0=^$(?OjOlX5@hhbN-9?p&
z5E#5>G0dv&>m{_|!V5{F4~)dHN7#;4wQqH-P+kl)(#uTj8<<@YMinVo_CWMVgZC_L
z3@JR_tF7~~?AbF0vUU6y0t8*Yj$G#3qwrE}`iQ2|ld<N?zsvxr2L_@K(PjT_PHtdg
zSV<OaxH@*!Bai_}?BJq%_Vr(raT@CIp7i8Xry;ynf$?VG%Ca{Jg4q>DYQaP7^sk2T
zUq|lA$)kTs9MoFIgK%pw6DR@BH(58N4@0tc3gNkyDN{^H9WQr=_JV~aG2ifHx$tUo
znx{Q&pyn(b4C-_-NI}tWdiB5cNbAw9zmlD-8y;+Pg@Na`v&9g7YxZHEj^PQVlq(ck
zn2~D?7#qM3=x<6zBq`r2Iqq*$au!mcTa1vf{>C%eC`DNZ*zF>b9TyFzz&MJ{2xwc_
zKaX+oy^m(OWeFyKpXKJ)l3zi<|NCL?+ugAKE0D9Yc**?&HD0SfKHSLm7id?6n}K%e
zuxJiiXgPGI2kg6pgl^pI<9+=jwex>*_}%27gy_F{)?l0tI{Y*H&YnDhi_}MeXR=Zp
z9%j9ml2g!c7Y+~epBOOK1%JnIcv$?#B(p)k3pt7{de92HUcI@Pbp`H0AsmVcl=PZi
z&LL930DJ@1a;5%rVbXpWyz(geKHTtX_RL#`lVHmQP0!SSSjedGDHWks(1>zDfx&k#
zjowJ!DN!iOcsujxHRza&HbcM1Kqp;c^f(lZyd8YRpC~^nP5{~uvwy#*A=P?#3blKB
z`wwyZ&m``@2c}@Cb-iAY_hxj<<y`^YoB+8D1tnbK!hnLaZ&;WF+`uTM>sQF^e6TBI
z08t2Q{=}c)KA=1bmle7Q;)Dmti_GxhdtzX?4RV@(F-Pw`4#+PCxo*>2mR7P+0!5KO
z0I&1AtNibh%l{77zp4BGeKz`k=m0i1%<tr)7cGBQ+I_+%Ax|tDyfG_y!@~TBS_+7A
zFuSpP1LMUo)4vNgNjlUZ&v-9@CcDMaO{TAYx^XZ=OntDL4x%QQoa<?ezTEX@FLcMi
zB1W{X-4Ao**6MWXcmk<GFf5n%G2JlBY)Sy5hQE#q{wwiab+&l;e-d9v!~CcGn#mwt
z=PV#44CQLn`Z#12Khzve1<PO>O1L)lb?k-o7cGzLefdEJ3FgjDY9jN}j>gf$A7R*W
zXJO)E$3DD==}y5S>aMDVl!Wl^ZEqu4_^fgNC+pBY{zn#$8mq0mQ^!65@oPa1wnV>E
zRZGnv>kIz@qHp+F#9U|+KNl7Lu@RImF!UV-+Mmx_|I=HG(cyRFa>FzdnX^eYhJSl&
z|H$GsnR0lT{=`(dAq@FKqhKwd3*5bo9Ae)+d0|vQuY8b466UZ#fy}Ox$7Y8y^s^aw
zU>gG~klU-V!5q{J30T_j;0Rx!H*?i+0euo~C(P;TVB<}uk*La_pf)%=D*6dDkeH6x
zup5}JA_qM0K`^@|{P2Ljc!54Uu0~oSqz($Vz+9%tcHX^UaCc)ebj>gj8UMBM_%F|Z
z{)bB4|7v3Y-^3DVg(6&QuZq6tR2UIR)d6)Tl)#5>SwZ>x-8(-%u5T-A^t`1PFCvaL
zQ7pM1>^lwKwW37{MxE{?S|YQIiqIHFHb|<&(vK+$mN67QN2yZ#Cq$0eV;+fq?KrYu
z2x-SFcmJ>CoL#Fy@ut{g!NC#E?I!g==Yy`!&zGu>JIuSaznyy3-hT>SI)0*m>&2JI
zxk<5RKqOhYU`w*wYQRk9rX1JP5Pp>BUYZ)T|HVx5cBx+F!}JI+R~|~(OWkelUhMjh
z%pD`fM+kW_B}^~k4DHml;BSSRrAkb!P;NTOl7%m<66sp5C<ny+tpk1mvIOl1$m8sm
zdL3Z0pykg)8lcMMf&s`!VDM4Ujs0(aJVcjLVCWz4O``*cgLnEdr`S;1u;2<?ObO?2
z6?#P-iwXNC9xIGsC}4ZytKWXN1V(HPppJxfTn}I296ia73V;P4fWLmnHhJQci7<K<
zzIljFQAeo508^rgnWbNsVRTE-c%22<J_Y)j7OmJuPd--EgLNE)74wWXp+1@HcTa`C
z1`kN3KgFK=$C`A_7XLRb2*fh|x6%6FCBt6-b}5Fb`K?nC$fDZyV8P@iH}aoJlB=@1
z>k{=Gr<NY&><_A6rIJyJ=W>m|^`z~I`^r&QTkD!Tx>*>ePMci)HtJJ5(jxkGL0A$E
z4r(}{`F&>Me3g=3OP#(NV*Fd>Oi5gTVuY~eN@*1=N$XuVv2NIBNwEuFy0VhljxHVX
zw-cMpnbng+RH^Aa{veO&S}84oT_^t3k>@BHnR1*jysSPv?liku{qBsLOy|Vy#hgI9
z#7X<4wx<Idv{-9o>brnD%IZ25uZ;MX{092L1ZwO9Qy<o#y5yL`uJ)hHf`OBh&PFb4
z*=H~Ae4JT{tDeixa1dzUE(HNIXg`5CrlwcLizBVzVsNS(wVUM|b5hhgT{21cFm5=l
z+`ViowexdFr@?aJN@i}nVf%1L=k*H$ft%--7DUvck;eubet+j4aJkO6=2{j?5NbD*
zIV03=klKf~xis!+N_R0h)msitybwIMQ?6{U<hy0A&cxkS%_TUEnK<z(oop!{WpY{G
zlV$QA=Fg;k7{YM=%~hXOOZEFj+i3=@`lnWm0y9EiWvvIktlL7_oS^;?*$E*aj;Vla
zqk^W}ys2!r&oSv=TruMRz-*n`IL&q1wOgG*MrVwimi3y8>L-j=Ww*ifjO#XBde1g-
zf|&r`fV%UX&E`bN2wk%&;|@*k95TDI`gD|E#LoEWO>>!NN3xl3Cq!6vfX%=T%?nf&
z<~q;QF72~PEzT3|F){~AWL>rw9smUrn2@bX0O4tDffVRaIYD(V7*2cV?ZZf@e5gqO
zhH2sC%gQ|WqK_Yyv3}pKbieASn&cK1{5kou?3C0O0-cWl!Ekt=rNw2wMwM|J@PbA_
zLguxUM85$dK<{zbL8MgTP3PMI?m)H`!)B0MN~{^#j5UfRuzc>~rWlQv<@Sa<>+_o`
zx@D<0gGiya>u1nIOn5xnMN4WL@qBK(T4az>k93|iM^h1*N%uypZe<NnOmYpw0zUEu
zn>S$3RH+@(Imb%q=rlWfS8MJ~N@^DQIoyl>_wdHuNsm6;oa^)#15>@_(Vq`!C+0D?
zRL-mzjeP7hEuC5M^f7Elqe4{@N;xb*X9aKh<q%fESdhcNd{S4irt|<Irqe*p$skvy
zqI{;rkkI*&sS}iz<8ejDn!)=<gDb0U*82%AoiCmTNnFskBtOeBU#>5|3-}&mWCL9P
z@Bh#>*k;W!?2ODTEHBW3*$>p@vaw)0Yh8S>*wj8a-HE0CpFAsiE@5HCuJl3YFsNg}
zYkwxS)T;#MOZZlgv!fIvN3<f7yH3YMjH#F0Ay+Qdo0QglD1R)Vff;QRM|Yo=DX%{s
z!Mky*Mdt(0>LGAI$v(oI+v16xv31JYdOr25iQzVXEF<jUnMgpSj&@T08^=!U0%;+O
zDFpUcb$f0~Vc1TB8bM{^Otf&loCYI5>|qH75w~e$#X9q=P9E>`T=T;|@;VzzF%QmZ
zDWH9g{M*U|TSZ>5JQ7o!JH(Ql6S4Eoy;+QUBL=RjM#tFk4LXZE-4Oqzdvl?#10n0|
z`ZH>f;ejIiDl~}%JkF__vl!L>_FKt<$*tAoPUj<Zyd|uwj3E4AV}_O6a?p`9){>5F
zRF3nT);mW~6tvZhO>b_yY;i?C`yCY8Edjjody~JWWyf&dEk}oinAQ`*nylx1Wa(0B
z!#SLG81UKT2+qDNMDFVOGg6(|lP8kOq=JlMWm(|5gRDZnAdJhvjhbu8N*fMgc|Q^_
zZ(Gn!l~UW(OTE7_62{JBMo%(fYZc;=S&D0FdaT=rzzYRB)mxLab0|0&LNoXeA4gQK
zhQ4rJjSF!t7YX`SA5tjjA&QP=2T`3Ti3$pnsF6wcQ$7Uq2k$SvE{lb=3y;a*8cp8+
z9`)KnvDtn;0_JjTnK~Zkw6e_De(v!?4|s)9Q*xp=u2ry1$q1v}@XJaluGH0d!Z>};
z^f}z%T{lfI%LUE$s_>DNyl+oD!MGtFOqqOsenr4WPp>?brIlyQ;(Zgi`Tl{mhs1>^
zf;=IqENUGKG0Qb_^G`zJz_A4(3V~yH8v=F2n2kT&%Sfqi8asy>%~w65PfZV0!I8h&
z-OCirmQr}~+%4Xct1&!0_8{yFcRZ}C#OmuyLM#XDOXH0Vv$5q*9?##t<5L_cw;5#@
zLjO?oGP*JsJ)CBrxLV2pIH-gf+y+tcrSaW}%FDcB5By1)z%!(yTA25KiS^zg&*73k
zTl^J&1B5Y6%zOH^$<7xv6;-NHfM;bk+Q-$RJFV#{aIS;+*=(%7?zCEzHU>DSAGkfa
n8hS=P<cReD5B`xMd54ww_2W^KV1@?(_`0NZRkKjz_QU@Na;$Ho
literal 15736
zcmd^m2UJsC*XB(^M+BtzAPOi=dWWbWMY@1=6zK>^?+_ax2m}yNYNR(sDFH-4MY{B&
z5Q_8;LMVnhH~7Bq%(rG{&04c&&3ym-F}Wf4+_U#S`|SPfa<9+ybkwLQSSbJipwduR
zF#rG}BmlrZk-@+-fzP(0002U%r)8)L{saIy0I&z-<N!JFhyv{G0ekR>0ndV`2&V}z
zdpS9joV~rAJ$S^(p-^%ttUX~GcnSc=zyW(X6k(k30{_6+qfquJ@QAg?VC*sA@dPCY
zpg=120P6HoaF{^m^b3W;pio#827|(Y#}gD5i^85@>;a6u90mox?ZIJij6mu13xmOe
z|1enac!D`O85kJAqU5kBdn^WgqX;A57)ayv3yVDguL2L)lamtw0O_7!P$yXMjX6CB
z5+IDcbCxRul|lYM-P8*JI3friNZj0}9RR=q4V9~g{^@JuF88o)^)EeolXaU--ma-v
zjXJXBCD6GiCcgF3Qq{h#w3RGS8!zQ!V%fHFq7n7k2RNht4S7f{6Bz1loUcB8e%f@5
zMF{wm*tSOo_<tay0o?yne|nw$X83)~6t?|){oA^SHcB%?PYM(zD;M0AMyYsNdio07
zbTd^``zHLX1t$yaQY*jf49Rc!vZaltnQ1?(!sBo5?3Nw3O0RvEq%QCLs=MmYo6LhK
zVSrp#W@#>8W<;;}^)-wSbY4mSai&cX(V-KMKD#j>ryT+{Y*@MhGdz51O#{D6<XAO5
zgI_H;e65iMjE6;9|4>BWDx4W7Ld}m-EgAt;eXF{pjSAb;F<5Tn5--fKw8ywL1|Wx@
za`L<tYGVm~5!@k1>Ui8>^^|Fxa!?w9-+mLEb#Y*p6Mm#E&4Tv)B5(_VB%(nNbpUJ9
zoFl=={-ig3n<CN8P57GHmx6X&==(w&QX`l8kE+J=i`0-(PN}*}AGfYR&LNP-0$zz_
zL0T&=WX7e*C8l<t_DIbx$UQjQdp@fz_GZ&AX<@CE#wqcBfFmiZ#kRyz_W~lWYC3nO
zKj8H$9okz#zTqX{!$tuq=0=p{Q69OsHus~IUiBQCQ1*6_R`q3b@n_j}_H;%3&dZUu
z?jS8G(QV(zQLKs|x>DlcmMtPky`Vv&Ru+rPb4WdlTkyBYY^5Gfc|ET_9Fa$?9&Nc3
z@-{5-e%9KY`>nM6U5~bu&nr#88_S~Ndts>cEjibZOUsu;yVvvgDR-s}nT~b=$%TUp
zA<GJQRVll6!*!z*iLPfNnfSw5b4f(kYobWtO|oz#-$rzvN{GGnVO2tS-Ih9DXD+8c
z%=*y2(O-|cH>5Uk<Uz2DLG%|JslcO!&eZ^ACTw^gzq}={<?o)Yy)xnJ`SF8;uooEk
zE8C-x4x#?vu1YNVMIr3{cJE&D$Y)i?V<_|%M#g`eOIa<_z?*_~VcQkHGL6jf@k&Wu
z&#YhcjN!9M+pfF-XwgIc7uUTF1T&Wlj+`1C`R8=F*K)FgE3M;7;=aRjMY_AXklvJ~
zb{S(5y|eGVSM*};+p&nn=H_$4--M1HeA}kg;SwiqlR4Y&?t8Yl%$H1WZ=^mjd4#l0
zV*3IjXf8+O*eX_C>ZUob)d|satK@s{m&+K)pp|BfXj2S-F4wnPiijQ9kAoJqMBQws
zT4!@;^2SzCuE5F56)x=2vY0KWhBgnPxsQ7-#F;1EyNMn?BV))^ysU0c)Mn$$mj7<|
z+6uq|YTWGFLD+R%QrCc>uX)5d_$h~3<+_o^tfSnu3=O7^_))_~%}n|JYOTTI<DHsd
zU%N@?4>{5k`#n0+rE7z3tI+-nB1U`y1tq^?R4o+ILf3_OOu0<#yvmpZ@k&dH$3n+T
zSGn$mq>i?7AH=XOrTNX3JDusx-cf*ej#xUU5&O!&j=?EX_F}CWagFu^4PLLPc|%t0
z=(3*sG>A4}<xaJou=QN&VsyBfQbD_erpx2d8vY{`AD8SA*$1n0uC@KjGx<3eZbEbG
zB?N3YdFLiBtZ5LW_)!dKSMh6QhV5O)ER2CWDx*~;n~==Ql5U3-bnq*3Wi%mfQ<&tl
z(=X4esvf?rvqD$a6@Hjh_vSh|4a7`{Qg44#Mq63zU?;EQNA@d*EYFZjy;(Uzx5b6z
z+HX0RjCKwq=dBwju|pFUlzQ*43`u9C5*A+o#IV-AOM{3#e~i(-%|coSjC8MGTM>L`
z4ioG4JGyiA-VAvCy=j2}jg$9(t#AvlYYj*Y%XH4#TYi)#e6@Lk_-OF8J#t%BFhFw+
z*C2?<yMfdZXzY_Z<pC~j?tm9unW?o6fzfpz4UyJSVFae4&ThZ5peB>lkXmn-anwwA
zbHpj*cZYi3Oy%F4sP(rdy|DtvUtipGSi_mKYYr)!&jXTvh+!p14yZ&UexP&9H0h*E
zYS`k;A1yBBlxEx?AVWE^7n{GD6MRpJT3Z#a?1mNqMhkUH$~&}0q=llJLwBW*1&99l
zYFK2s(F=|JXGO$?hlBJdrGmXJL!8%<yJ?Tzlu3fZLQOaYFTo6N<YIoX-Y^(g;{P$d
znW6Ln&mqqW5sSI*TG;Hu4z(4n!{*l>7Wg1jarLz5)wNNm;qyB_Kav(C{+Q|}x26(j
zqmsI^EC=jG7-kGQs6f5AKGb?R5NM=mU`|qKkvvR;-VZ=T<7<`1RzKfBYR$kwGgx&?
zvtIyS%^&Q;Y*{stWWCPK5svv`>>c$HH>=6zhzfm(3w)nncukV*hPN$X-3Zykb5(^P
z-ayK+L@u&)fv<(n`I~-=8sDFNL!o!fhU0zN(AspD?!Fyz-~AEGLeO_WlD9!@cofV0
zN6sy714?_3bD8^UEbxKuuIbG@h*?OjoomEf<6Q477f+eu2D;J8U)3?W`M9b(&#)%}
z%t<2e$n=zma5{@akn)Uip@D)fXbJG@FWPo|l>z*a^QXZE^K!7D<V<h>%$H8s;8vf=
zP+7R{<4W_86k&*`%$c$^tzqA5lj&T?%ZXC=`uR+V3O&Eo#=l6W%m2U`Zq10^VGoz?
z*?S9d)(XBTGI!S>rw3I5!wb|feWtF_G4pA1h(T4o6bCeS!!|R<ipyFm`O&h5q+c`n
z$Nk~7-99(#-{v<AZc+9c87dX|XLSg4vfN{UADtswqyI9nrH`efP3c}YzzV)2*Xs=4
zU`OwNU11v;`MQBQdl3N!Sx?zcKWNhJ_)JD!$66`05JIFu&@HsRCnV&+RrDmFRJs0z
z5>Ehbbvk19ab~E>kkB$$cSpQg`LF8dx%r;#2lLkrtt}F>2KoYC?_SH>>(C|5*BvMZ
z^1gF8-sZ8**?nJg<0|L}`6UImKFCq~hapH-T)h{Pl}@=t*^H!6^b}k=06Y1I3Dhq_
zDO`@^uCEE2G@a8|2x|1`dOhVe?5n%jQWbO1oYV5O(e8p|Oo?nKUzPQ9vn*H)O#Y_G
zjUf>&ljFbXe=Y|kTfQqy){#2yfNMPVfe~~Gdl6!ipsO>-zF)M|)Q?f2@6$giof!YJ
zEm{&LaKp~s)mfxb$JZ43ji*SI$gvTxRAsyFg5y8fivnY}NO$#KC3jrtt#0EZ?c2H<
z5kS?qzb%1<*Gl>Qvd?k&{HfFBuPyWMLB8)&Ia`k#jX4$jJZp<i_^2B|`eu0JP%KQu
z#KZkfpGE-7NIAfwW&pY~KIrqh^2?SRY55EgIic<cmnIv77Y7W}-i)Tb%}fx6i1qx<
z=@oCN)wkZG=~&9YSP>U`P{(@8&mSLb8)v07^Q_%e38d`)K5{W*N&A}G980*i^dz5j
z-Dt_!O`p&_PD4LM$*h5D$MqIRFbJMQ_X!*^z8~6aZ!D-1YnhV(-DsyJuPCA-Wkj$S
zmRk?z21{Vb5`0IA@@Vioi*lM~l5V$c5&w7w<w3A};3M~c>d&Xdy`HdrRxt=PQvpi!
ziS7T`XIAjsOpxY7ZEi&D?ACQ8jWBeUKkB%e2KwXyEnJO=Y%vayJ&!<y5w+1mp!+<C
zSTdSkE{LZJGP&;fmsnIbn58qujm6|xc1pi$jLSS4JX#vC1uFPQtJ!s&+SQdfRYV$C
zIOEi|27Rx5k8qx64>R9Yw=q-X{Lr}FzS(9T_spF@+HBt0LQp&>arJ!+>7rY@f~gz=
z|DZ0KPasXYAVy8eD9f2b4>#9Lb`HIcdGNlU7VV3x=rBkW!la$mYv4W4iLN;()ObYz
zdT#Esqmn^`4-evvZl;7)Juw-I2Rf)DM4nS!y{SvRFq4e{XJkU~Ky%}J78~}Y_|f3~
zgSz11!fzk(<tAgTu1*<0e{0murxe(C1-n9^vyniRLG%DNxLtiXJfYdE33AyN*+$Br
zX%iju&Of`H>dkMC#H>NgEvb&g;rJ}OTBnAO%!*LkdsxO2R_Ndj^Yn8A_j_8xVG`}Q
zpWF!J62vGx-$~Sdu2}0(Xa16~^FVOGwXS02b8t#?jL8r!T54c9`D%I3H6gL~1h23{
zu7+3EYZoKDLY)Fi#1OdIPdz1Z&1{o=C&vbc+@X7=kbFQfy#x|tmDf6(z|>V{b)zpf
z@XlKdT~@(C2Nh3k%?TRA-<Sj`ZZMbY8oK|jxXx|4F4-%wlKGbFb3P$msW+_Jl|Kua
zwaHOWh0IHr$@D0P(w1##L}Hvb4Lf(oO|0P?QTObGR;X2x>yvi6WB7z)-bvYmmB%*H
znhif$p@!=n+L@&WTVaz6*SjZ!q}lRD)#pV*H{)QgaoTo!A2J(l*yV`$t;b&tU$gOF
z7Ic0ay5w@Ros0&~hOIJ@q(L)oX6s=i+c2Ed)`y)B8&s4yUqlxMd~S#3DvF1=hd@M9
z6NJ!Rz*0oQZUx^7H&<eqz}d?g!eFA(@Y}ll>fog7dq=lSA9B6&X`fW8?3bZ99kgd_
z+|=CC#>^DmA1+|g752ew|83XFZHE<=_Nb?3kM;S;i-uzS14e2sa$7^c1>N2PwEJPo
z?P3L1MUP%I-r6^CWyqz4@2plk3vBSpyS>=fqS{w-iU}-O@pzwM!qMJ>X5C13#CkeQ
zYJBldi1TPLFKB=XiN`qC<!`4jJ?2A5k)aj<**yn4<Z_+Iz)4AoUEL4soaB^ezO2s!
zdR%s+SMhTK1*7r|B^UAK-^ui3Z;Xt-GV6%GzB{M89MPgKMKa`H{bZ*+i+&cbu{G=*
ztP`~-N>mv5vE+EmCV?D91)Y^eu=W4Ws$xV3U()2#>)cJr+eji-59-`bnw$9kk$RN1
zEg|R5h*jd&iz(0c?ocBx4OeVIgI9eyb<1uq_HZvKV(yZ`Wj^OOnrrJ)_4+)^Ln+Ff
z5|=+pp1O_A2rCMcW-pD^7X!<y-HONxz)3AAA;)kuA*;Tq;q1`xqlmJ}*3SEk=+%*!
ztYw}NcEg$Rj?MBs->?mPSPutF%fvg<!mTY0DNdrc7GEJ_Dx$)tQ~q5SCP`n2<c!26
z26MS$J$_dT>~EzQBCvC5M>XdWxRs_uBz74vx=}7$*iyH2E4S-OlKg9?Q3{jL3vS*v
z9S)KqF)2*r+z2U*%X?=yLMWMocp!Q00^5Q>Y!egY@`Pv+5`*xi&<m)U6hq`+m7LJd
z%C=|)k{p1G4i{slyRO_)$#=!^;Z<cj{t>IWABQYg>)<L=78K|W97gt2r7x5<987VM
z9(_l$ZWyO!2eZo&7qZ-JnyO3|dUrTiwPF;MPO=UoVBx5sEZv6>)cm+#GZBSzGQnQp
zUc|o2gQPMph$#7z_eO;vV*EP8oEoXy*gIPwpD{vLpo5*sLdHHsg)wHTy-YXO%-#i$
zXhvnL=B!Z^Cg<E4)ijO(j}$5xVT=M|C%*J-G2X4^p|F1auk%=Wd##Eo;_wtYxbz>)
zFkZXe<Usjdu%ido!I;bdaaN`-l*w^u4?x(74F*J%zzjLs_tB{v$+}+O$3;ftkDd0<
z+Rd$N@6ngIB}rOa-QhyTxSE$blj+sXh9X+S4<M?2Y1SRMS3|K0)9E%dB+r+$7K4?w
zLXWqW#b5I+B@S3lI8Ry9qC3=iAAcAm*y(-djGiaPwJb%q$Pz0;6ZA9k9VC0oTEbcM
z<+TTXZLvej-}^)xF+j|+*J2}^?TnYsNnXlIwB9nZd~R7r9DUi{BH)e>(wWEfqg>s3
z)uvnY^3U6K-`_sOc3E3xD1{ggPS)X1sf0uu5rg4k-?H}k^4^o~vv(bWJUY{&*Dp|-
zyCCnEg_n^OE>11Vj1QZ7(3i?3;E&!!mrSi}BvTF8e7`9w*_`ZrZ@<#C?+594Fh2X`
zC8EL#A&XLPBF9=bT(wQ%Q?<IGZm=8Ty-<yJqSo_JMo-@!IYLI|=q4RNhM<O_(JCL}
z{D$uS9#x-c<tdt^F5GqN$SqH*G}~u%@1WG5Ul0)(xJ{)OQc=3OMGr1|v)MWHU}t)I
z8uzB_=t~_B;<27^23>Rv+Na{<B?-;fB#ZNpcf&qQr>t-RkgY-m(L#lUdECwSrR(HW
zmM<Zfs}lqcn1`PM0lBeJMQ@TKD%6$Ft-CNpGQj<=k~k)>N{z{D)@$uDzdbmKC2He2
z@+`IKixdCp#L#BFkFkz~7?4V;aaEDN-C^S^C5LP+b$BzB>`1?O+dlqd%Po@sDKmPX
zJo}CGht<bGzz&LCvN$BGWpF95w85f*jk6Yj4Sr<D9_B%hXyE8Z0A~B)UE$8bL^>@T
ztxO1wkRxJd_`p%x4<cAN|0kfRLgmb)R@MAbBA7Ucd<8B}{>-)J%W1&7jjL&r)AWM!
zXo@6pavVK)efTo-S<U(_Bf?n}q`&)Cr+dbO@_kibFu$;V3F&YAby0I8zuJvJs26CS
zmu6Z}Q&VeYN@1(ji2(UNWkXZ(AlUTkjyB~A>z=9RwA9yqYjjo+Q=Sx`*r!~%hGeHj
zQ<;Pzx9T6}O#1vbnu+(cp2k+iGElq*dvrU~UY$&Fd)JXtch(^NZavO53h#DHT6p%m
zEJG~j_vxQ7qa&~ivLp3<YVITC^$f!t=);wu!M$T<H<aEHNbHa;AVIBWrp2C!89n`>
zEC9cD#ifyp1v2`?hqUmw@l~J8!7~S`E}%pP6JDV(hkhOgPh`SIYfy-{x1$+sBN}l1
zlnTqWWH32#OPeoCjdQftm^|KDUF7=-T(ti9D%g22k9mw%Uwyw>D#E#`x&h}pYLU&A
z(4>{Kdf*7=^G%|S20h;zKNsj7oy%NKPS{Nf-4S?^)@Zh6+bP2N<S7y+ap`tin#)hH
zkY*gceK53_<~aEvsmjsp1sFL~hze_{wsm$;zsrHHN~!`4pNlp4l8d^$h%`Dk<Ynm_
zv%c&$(*}nTd-?2!4Gve1oQm`Ag}oO<jJhtCtdxV*;-S04D4E{c?VFhvZ|H_EjR8jM
zS22g-w)z7dufd?Mae2@=O72-cW{{E6_k(cmS@DZlr{ycRFkCI{=vm~TS{27|@PO~Y
z^2wv!lw;@i#5C)XC?9SF<vpZvEC<9ne(cP}N#|lAOtX|pND1GusxhY*vP04&jj>b!
zfg=|{q@8m^PE!ZqHM*33X3)YN%zSxw6x}^c6JjNf!}Q#w(qOqqLMtwOH%l)eij{+L
zOW{PU@ExZiWmU`pk_;uj7U|>F(=I`!;R6Z#y&tmM?CJ{X--EaWmEShiMb5L;WN>)i
zCE?#atWF8zSCSr{q`ZMlcY3HEs&&FXay<2$cE^w3*m3@Q1b$+U0n)#8B4Uc63lxHU
zSdduMrjSC6(k9}2h9{qEBGZjv+tcP35ld1g@2};AHE)#z<3{dRzDpujWGM1?12^6s
zRf$4K)W<0n6%M#uJhbEHPMXM26?ra%DGj9#E5)agsB;e@Bug96v;ZlhBbIt_k=yes
z34gpHKegW4YE=?p!U1y7T{l|jos);zR8&T9to6$9iI$^}<kW`^-m{i9kjq^s_94ob
zptD-?KYz%)mNCWn;Z!Z7uccoWLX2K5+osSTsCM=1#hJacU0gjLDJ`IZlM=O!Sb7wr
zH^(BJwpYXKBrh@H4;}|1%d7Po^@JjC{#w0xOfvpmW8ac0%QilU+l>(-b{^54z|rE~
zP?=Ggekixne_@>HX5&>jRxRW3G8UKcQPqQYd-1~Y)YLcoybFyvy4WM6FxwvNL+J!{
z<W3gY#=Jq)W@_<rVgL0GYgJ&ppOf)qNIGeu0t!D_-X(cd#3*maDlIFRQ@$S?xu4{G
zKiX9t4N1-7TCYF$s_k>HV@d>MZ$&hetB&~p4m1<qj6RI@$eHox)AguY7QFc*^u3F}
zG5fRX(MyM3Yf?R<X4*%$chu(b7psUIAAdQcx2tivk@`9COe3~_vN=t(r^}kebVS_y
zjT_^h!m)TB>9>x~Ij-)O<$2}N_eHCemt%!Cepw!5Q-)MHH&@#Xc86X-u+iEgo&8>!
z(!j4Wrf34(B<gDmV|f3dAmhxhn#iv-a1%07MDwPh4@-|u=#$4lI~?(j<_@y5Vt6oa
z{y@1DfdnDegzO3;@c!1uEpCJj8I2$jC;IRPJCXVs$fKvduovU#H~?~;{tnXlwlFW#
zHVynvB=ihQjsjzTv``Vohxi0MWrq83_Oh)#8PV|6IS+x7@PUZzaBZbtmIv_(PRs)L
z;hEHQe|z7ZLK_K?-z9MrZ9IPK$uo%`c?m!s(>Wsd&x5xsptdUDC{Y^-_UFh@U=etM
z2F?Q|Qip@PM212jpM#nFLx6=Aj(`$&{y*k~a0a;OxeCVKf0ounf%<{GU@wH`{@+_B
z%>(3owY7+_u+n6xm%k9GwI{6TH0sfdI_|=*);g#9UOcv3yWE+f>+6mHe}mCn%D8U?
zjOK<X%o%3&Fv1*Wbn9ao8AAM?fh!)g&kKRd=Kcz3=LEVLChWRSJOVcbdZza<gnmf6
z#$JE>%e*5qI1h|T{yqio>+J&Xv;4DFpnXkiztcY|lux^pL1HVN^Kb&3YSUjBGb&W+
zf@1`Clg0+Q-s17F!I65b^~kA|UPu)b3^ZPjh`4C4?WU(&_X2sdhxGlSM3H1-i0BfJ
zHG}7U&&Jtk;D~E4A!grtc8FqPUCSc*5XdJ87=7BLMLHJ;2ZyzV+*sUWP}DC}dLsWA
z4fjvLLrrA`BKz$UI!+Jvm4nujWLByS;&7KFQvUo0sy3fcSnQ)SG=3`H>?j;N>DUuy
z_<H1`9DK!;YMagmS;LdP0ygy}dT%2k)3K4=u=2D%ggb<!?Ks>v{&VqbFk?}MR>Bd@
z$o=lXPX_;b5y(M~y3O+jEClqD|Bh<<g3M3eHiCGqklJqdPx>zj9Nb`@=~!e50#^Am
zHc}Yo`Kkv&Jw;N;^pA%BI0JVt(A7h|XLV9LhOam@c+LxslZhjg<6P7l$@B^!wOy{C
zOi^HMK>8DRQ;cv%7Gz-rrfMlRtHB@Z6>ptrS6oNRfeMI)#a=tpD`Ab(-^<sfR3V)6
z2#37|{oGegW#?Z1dFY@M!cCn00plIy_K(@UDL5j7WLBaP;`fl0zmY=`vMd2Fz*E&w
zDk7ZYwioFj%Tn-^5vpy_6r3lcM5D7EtWWuWa|OW!mK*zM&G-brI!UgV0vYLXL8s8b
za=T%HxSRNmQv@+ngh#xh+NQNZCWyb@0kc3fsFIi85nNZn;FVC-pC*76i6HF#6<Fm<
z>#*Ck)YFIewUEZtZGlR#BO+FG!@mPooVkR+^>E&3dJY<7I6dfKi$i%4qM~q6?uW$u
zyY^&MAc{TVCU*2V(qZ6dt;e5NmNNpa_DUfj#3-@)<_7?<lM<c<z~NGy9><2ob7oP1
zTg+^NWVCyn$ra`j5DpN2TOn=0epBvMd9Dyi0(g?JpjG?O@dZP0!l)zit+L9iKE5K_
zeh@mp(7?PyUb-U3gMS&F4qG70eX{OjU5=Had@r_Mz7G0f6Gg8!X3i*+%v(A(^XP(%
zF~PN&4UqlfuV=K&mhTa%-$xKI?2Hg%r^#ybo`?cF84tLUyTok|s9N59t&srT^a~0q
z0sqGZ?>>N}o^VxlE^I2FjLan60S83i0pncxr1hNC*?k*N)YR4faK6vZ0Ex;DS<=5u
zkQNQ~X4-SumKxa&l)?G1m=IuygaNIQo3Smv=|y`Z>W^yiK_Pj3RvcR%6gGdIS+--!
z6#k<APp|;htUewzQ~CS5pbeei%A-G+%VL0JytUSrB6og%Lq-9_von_w?IX(WN)r<k
zp+?dq%eCM=tD`^<k4^mWoF!;ecl98e#zEOtV^sTKHKns_jnVug_H@H)OB&~j^4>;v
zF-PZ1g6yB5^zJetN5_kci*;6$WKuVcb9ZA6+V|LFb#6Cmamhl7$e1mV!|>XQ>`?3P
zD~IPQy8m0yEj?QVQn`kN6ldzZC|4{Z>})tA+Ll^xUA}<hYW}T4SKSicQ?(a*GAnt0
zq<;*^Qn$I?fYno7MpPZkkQ4^$heT<=)A@r}k12X3V!KPY+$bAs>T$XHY%A{mgz2FD
zSy~g78RuFpPG0dnnb`g4{AardGZw#lG7UR<E>HDQGj+r*y^8c@!kjx@S!KRFU^(Q=
ztC~aC84vC~lAD-+g2?>4P}}@MQm0v*Yup0esSV^xWvmL1nAwMVd4zg<dk1?gJ&+=V
zc2Ln)8(++>g4%mbfszZAlxtw%N+KW_XlMy+=XaH9-`jx2Q$^E~3TP}~yx@!yXD%CH
zOSGHBIE#A{k<ox|!m(~tzw9UU<FHiWtbMIl+8=9>yabidp9=u%pTct>*vp@3^N0np
z>};=Kx6X&PJmF>#RXIpzWm+M(A*7B5R;KxH#b#n7OGszM>K<~2God4*qkAS9y;=pP
zm{k8NQT{@^>y*7io!g7wXFhi`&CnTlLJeh0Jn)xd{|W<*4~Z8Al@BLoI{cxt3h-){
zvodVP$iWZ^-(w&4>T9kD*-%NZTIxcrALB30%YB(qovS2)%!bIpjITP=ND1F56Mcxm
z+$$&f%n?!$aD^+_76}#?XPyT!=z376XcKsC2RCVT{Mv+gI_NVPxtu~_>manOx^0;v
zr-!(!U9TKP#m;8TE91Hf$4UOIhaPRS;c+gSb8r47c(X3H6Ri4%?pLc3l=bM0-uWWE
zV;xNEHaTeA>JYO5UK3mhn7eD9Zqg-$Gv9V7EqW=iA*Yq<&$d<YE0XFHJKBd3AISZ2
z(cJDMmJmbG!6iMFXgxUoxJmH@orj{e!D2XJPZ-d)9Q-rEqRVmwWcUQFG#Lxi``#2c
zVbKw*eM(mt(L~5iAS8vGbF62r-xCajs7>K?pL2n(M&;uD?lqaiBYEho6ue-Bs)o$~
zIlS9dIJ}2;k7b8IiHKOxoPOe+YOw$W3TmXe$#uPzx^Hn7vnPG8CXR3gXmTIwJiPm7
z#vP8Nz7V8*2QkwMvW%v6f3jmrOu*tP5`HV<WiR>Ix}2QGn#NRu>}=?UI*&l^)6urw
zl8?iu2*ZXh>S2Qt){lTmG#dF`m`N^mV){yQLDU7(g@PI?)Ar495PHvdKS5Y`WT^}J
z2&Qd_JiDh^7CYYacpc2C2|)n@68;}#zJ$<G#>AVYm>#PRYT`OYubkWb$F~Gm1E98?
zr`TaY$g*%A^!3M6)6=?}VUzF3lYOrK1<jWb(arbS>$^O~W3sEWXKwv*G`%@a+r-Sh
zq>ZN_cMwOPg&sLs@VK?c`~KH?LqTv)wXfNfHn4=3Rmkwugc}5N!y~!REPg8A>floF
zXQq^n;q-8y*%FhbTII!qosJN~zJg#a9h=>!wli?(yy`^^0@isF26)v--fEr$B8--+
z07|eR0YvnQhX-*BW!w?C_DC=i>h!uBKF)01e8mMNY6BrL{2?gap9<iN2!gtOXTvLN
z@)@`SmOt)y77<4(zxt&6Z&64X(G`b`wsE9Njm$|PQ_$R;EV44fI9kC>$L_&-n_=Sj
z5RGqU2=PLoNc^?#bP|`3;*FgzSDgL?TkZ2c*UKdk%cYCs^S7S-iCy$%+l5C;rpOv@
z)4soF=K^+qKnL_I*rgMWq*AvThO+*tS*}yiq}>s+q+vsSEhUz?PJtRFueT<0?9f^K
z_?H^(^&qLLOR|xd5zE5}_Hf;QL}!5etoJPDQ1l*_(ej;V{F}yO@UA6xs+OK-<L^ji
zj7;!BpJB@R=MorYuZ_paRZD8+K(9;kQj&zV)Bse3Y#y|92YIf!)~3^#0m6gIS8hVq
z!A;z@(#&+4V&nqNE(oHw!T-88eEoj*U*WZRNM;m2WgKjnN-$t`NNQNvcAENk_s94D
z?l5)|a}fCD0!G4L7g-4P#=(!QA%dxbVP?u|NZ>m#-dK`%zjf=E5<#F#2p4g-lEqby
zJS*PYn9=c3LXv*%3~KZR+tPxeb?en~Y&s!@pn*xmEZ6F5j8+Fwf(`@LrTJwN>82AC
zr#TcYy2|Rw-9T|Hgs@3Jq1z7*ZuUs?BL2$ALF-^0eyYVZR6#!Mgzyz-T@(U4h3<Pr
zLpRnK%Rh7wXn<MnxQs9=RBh>C4^v%3$RWYpGq$O<S$2Qw8eUY9u5Xzi%mh@3z^c_K
zWN7iL=PalM_%}dkej)JDjaJO<uV7@mhK59SxeIs4#m2^p2y?8r68!65QgpB`1M7wz
zB>OylZ|fbXM910GI9!&Yg2iY&nBjv-ecE?q#b=_nuK|{SiVzQ9*xSF05bP>RI@ok1
zKtOEToiW-zAH;A4>!JfQ#7yM0hUjFW+Ge;z;DAG}+Dr7BBO&FD+^?KY2L9ibBmdLd
z94!9i+^!Hx|Gy9&tO`=CJHX20Y+<VwsRdYk&h9Gv-fsL(NO({Xv5Axk3eakvu%IWb
zMRmb>55csb=tIv9Ldgl1xBfh(J(p?z{M|g>*JnLuO`!gnq!2`q|G)YLpzZxn_4@D8
zBzXe%fB(vnchj0&z&wI`5Eu`R)clwJgJBh}eVqzi4R;&R)pA)bNUbPknJed3i5~%&
zXkgp^G^7hTyPtg6a~+gNJ`xi)ml^G3*b}4K-_T_a`a75jr#8lA?~{BNjI#Q{4!Q^O
z2>_!nTbIC8YsCo>3YR~<Bj_2ukK|ItTs{tTr2~I^hnW^@m44k*-#?v<KPb@4wD$0z
zMmZ)aJC7jPSD<L-mxo7eXpLT|R*y=h90h24v|zWyL-MU<l&IYMAQ%CX#0iglWup7u
zuQ6;AGE+yS4K6Jjmy?q-6MZw{z=I%a4Wyigr8G1~kCBW9dIQNwRLG3_`i(^5_2~@(
zoXD=oAF$hgkjA$lS1&rUC?x+{o~kX>3RVxcRKPYr-ZX&y+j^)inCH@ZB6*7y%Hm?r
z8a=Avv55eyc#ls;pCGBtBO2B9r_KBKO-OdIlW_Sbpxr^KH`H7>M(eSIw6K@%y<P(;
z9%GKjdU&&<1`bKaz`Ge8ifv$21DgR|?w>eLJ84fIfMvAf9`J#U4t*AEH;f~wL61vv
z$Y1lf=OQ!`zPX_zuE6Mnj<KD}^d*R35tC}LyYS$X={tz#MXV(*3vWN@2J_5ZK&FA!
zy&JK2sEjb|MX*uDkkD(;AkMb|7ZWJvAfx#5Ej$B3X>LeOT%+kccz;DK6TE=}>g+#o
zPw(H~NB;*I=06Qca6A4__Es04k*LRh<)Sr*{{*_+8?7uXf*X$Jf0;p5kbi)oS80ua
z496<43leatT@<A_VsFFwq-h*)yGKIg*mT|VsPp`l|Gr5%`U{nQf+4#Z2?<n)e2mi%
z`YCs|zar~C*qU_pS04^Dd~)eu`jz0x<S>KZ@9{abCx-=PIp^5eX8*N)2~u1=&i!SB
z!;_04HkWnofsZr1l)6k(5`KQO6>?1t^jorD7e4{bX?Jq=_5HBgSAcWZ#>f?&2Elw{
zbo|FJK!!pT;Wtx*u&yLspE;H=jspCdY><N<ziVXFlr8v`^BvBcaXcj@IDpXGmMTlj
zoJ6X#RLC4c+bU~?oRy;j#>$-iFAYost~=;2R5ZxXJpWsdp><zq5C_Q2k&{u}O&eKM
zWZ+W+^gmOANMtm)b#5UOXYEok!BQl@Mado{TuaXf_?`hHl)vuL{Jii~MjN7Y4pPqa
zvlaeHgTf;fntlOlEAlDsDl(CNT7p<cgQ(34$@LiEyY$tG;z0r#svb}#f#?`S4$%L+
zKno9ogr0$>=h75h`lN9lu?sz9Ci;t~3UZbB--sdfS0xP`nE1c(x$lKsSAL%3xIG`e
zv1t<5`q)7^>11@@DJ^S5Re#$e>(QI6{FUpH8m*HZYM9Uj1HYNY%HI*I60RP|x6}CC
z=b;NhTElmbmHMHTI|IqO2>SXr%9U>h=ue)Olk+nUyJqEno4HDlPt9MW>J_}4{x)>C
zw~H^H5&b4h$`zVD-g4bsR$Dc!gZ#&?Mu&DKIVv;%4Hc?*SLb7@KTgNmv)X8DNtpNE
zN`cTTWT&beagJ{Db*zH+fWPM8@b}3Xa{u`IPhF2L*x8%8el2ivfk|{JlQ@rz?<q<t
zTJn|DH**EA6u26f?iF5A4;|Zeic*soDl)q<kcA0J5G$5dMS_#eVTPfcgPfI2lgY|K
zKh|eXuqy5o43N=dr?eL<8zc1hk0hG!Wk?zc<vK4DG~m22!-Z>1Ih1g9K~CRas8CCD
z!fL!3%W{-|eI2x1n`^20f@Kf25;~S7s8Od^btD~joVZu_dvEWih0+V=I&~`dL$q%x
zsJ2^Qey$O9+j|%6;mj5|MZ*m5zt*YVEz1hjJLxGR;eScE^0U^!VQ_j(C`}@L*y{9&
z@*?D1X1*&GO3BY2`;qcW0Wm*FaPJ3+n0DgqnOBLk3vgbigu{R+jrc5Hs07Vj67`IU
z7hFV?cT*DpMP&`xrYM*QjEF5HJZ0$_I~AkdC^K;q<R;<enp^CS9fpDxTDq?4nn
zfa855ZgBF=1P%>y`_%pl-a#`<AW@>|WaA;=!|=(v@$x5NoT=bygt)dQ#@gk8uV{n=
zI%q@_0r#HbAii~5a($N1k5o*%$Vp7pTv=nTHi@0Mg0V$|IrV2ZEVWo*D*BdF)64vC
zg4Dgfy8RNxTVmQ+Ry0EAy3kW!-fdGu)5p{{sh&<qQ0ZqN(^`vfp|vyxxy1ZWoYHO#
z#cd4h4j8UDeQ=}sa$yMewdzg=vF9bvtBu?vySnBdV}F$f-mG`c%J&Z#^eyBID+UeG
zptdfeWU8*iUAAIP`d<4I1T=Gjbrbyn$C83evU!M%hyEwrlQm=DEzcy#xh&8S^{!Os
zbQ-#~e^?tZwyf1PuUs4O4%5>?z@-BO9Z5fenT~^Ugva#>t)0wXtfK4w_)YLh2E2q_
zwW}sQ=^kifp8K?ZQ%<APz|*ZHq9SJ`CTNtCS_?9|?qtyK-b}U}tz?jNEA+SV89<i^
z0{scLnf}c`kI?uLC*0?gz1_6Kmv@-SXl5;rmUHk^bW5V25j1f1S6OSpq9^liA4TLZ
zGlO>mxN&<uq0R^6<NhgK??5E~JaQN^IPoG}<Hg3wtC*bXD(mj-@v0a`3>oB+cnP2#
z{S|zCmcQ85tJ%HT_{S|3<j-%sVEF1QSZ!^`y5lPLDIN4TgA}-l$hrlg^vXM`m=uTO
z$;P6vEQ}IFmp~*M=<0}h+j^3!pf|Z79&^4)_8J*G<Wa>sBr3}k`z7#Dq-rNaM!yuQ
zs0b420J_SkbNB;y-t)JGY;B0dkvbWzx9k{pg;`_g=#-P{5xUgi<nv(s&o&IrpOQG(
zE2{vs(@LZh=XKy`pPUle0-6OxY9-Vw3c*JkSX0LLT~_2D^4@=}copOT4lXfsgqeTq
zX?1ciL=ArE9g?9$1)?#&;x2>stRv~ekymKoGM!*!s(Ww3R+)j#V)@1Ooy7f((4)K^
zMWxM>ALQG!r&mY_RhRK<IPx~r?g*+@v=sF$$B*iuop9VlalmI&Ms&HfvysDX*9UW0
zgUv=9kTGnHZ*9530tpNxqe;($^jGI6{G4r_U~2E?x`!MLc;<KV{sUbkpr{Ip83i!N
z$0h!>Em3T4Omo$++WMVUT$fJXd%6YARO*vIHHE(wTA#gv%sdYza_9;{RJ7Gl7y8l*
zP6k+xlYiBJR}cvb(W!sMXY}q4|4TV)b@`1ebz)V%ikBB4r!0&!NprspDA0Y7c_=qR
z?wvBp9xwhni$E=3{56h&(}#E4CzRKF$LyWVt%Ie`#2xg}bXN`I>Jm5N!7$$J-SW_{
z#KVQ9#hK}2IM!^&`8UCyO!rErT=)d+UF0-swj_Dhekx|3CCGOGVQB1FCw-C5^5)w%
z$53;J9{bzrk}owp(}^m`cUM)B+pEkTx{C$RfV*oQYi~5(-;>qp>t|qJGZ+5__SZ@?
z<Syr9S~$BaZv~}F1I6{>*YD`kLcWY%oK$#uslpQ@C>kujOs9%`7kYcfK5QCV<!-CU
z`3|<|bRO{o8G%dj2+8C+F$nM;HAWx!xi9q$fVt&D;KtYAzsjCd&w&3^l14)L<R<Pj
nONp{B)4#pj|F2=G>xA@4Yi-#*+s04;d}*lasJy*)=h1%w%`uR1
diff --git a/public/search.json b/public/search.json
index 6ce4f11..e74a1fc 100644
--- a/public/search.json
+++ b/public/search.json
@@ -18,7 +18,7 @@
"href": "07-basic_statistics.html#cluster-analysis",
"title": "6Â Basic statistics for spatial analysis",
"section": "6.2 Cluster analysis",
- "text": "6.2 Cluster analysis\n\n6.2.1 General introduction\nWhy studying clusters in epidemiology? Cluster analysis help identifying unusual patterns that occurs during a given period of time. The underlying ultimate goal of such analysis is to explain the observation of such patterns. In epidemiology, we can distinguish two types of process that would explain heterogeneity in case distribution:\n\nThe 1st order effects are the spatial variations of cases distribution caused by underlying properties of environment or the population structure itself. In such process individual get infected independently from the rest of the population. Such process includes the infection through an environment at risk as, for example, air pollution, contaminated waters or soils and UV exposition. This effect assume that the observed pattern is caused by a difference in risk intensity.\nThe 2nd order effects describes process of spread, contagion and diffusion of diseases caused by interactions between individuals. This includes transmission of infectious disease by proximity, but also the transmission of non-infectious disease, for example, with the diffusion of social norms within networks. This effect assume that the observed pattern is caused by correlations or co-variations.\n\nNo statistical methods could distinguish between these competing processes since their outcome results in similar pattern of points. The cluster analysis help describing the magnitude and the location of pattern but in no way could answer the question of why such patterns occurs. It is therefore a step that help detecting cluster for description and surveillance purpose and rising hypothesis on the underlying process that will lead further investigations.\nKnowledge about the disease and its transmission process could orientate the choice of the methods of study. We presented in this brief tutorial two methods of cluster detection, the Moran’s I test that test for spatial independence (likely related to 2nd order effects) and the scan statistics that test for homogeneous distribution (likely related 1st order effects). It relies on epidemiologist to select the tools that best serve the studied question.\n\n\n\n\n\n\nStatistic tests and distributions\n\n\n\nIn statistics, problems are usually expressed by defining two hypotheses: the null hypothesis (H0), i.e., an a priori hypothesis of the studied phenomenon (e.g., the situation is a random) and the alternative hypothesis (H1), e.g., the situation is not random. The main principle is to measure how likely the observed situation belong to the ensemble of situation that are possible under the H0 hypothesis.\nIn mathematics, a probability distribution is a mathematical expression that represents what we would expect due to random chance. The choice of the probability distribution relies on the type of data you use (continuous, count, binary). In general, three distribution a used while studying disease rates, the Binomial, the Poisson and the Poisson-gamma mixture (also known as negative binomial) distributions.\nMany the statistical tests assume by default that data are normally distributed. It implies that your variable is continuous and that all data could easily be represented by two parameters, the mean and the variance, i.e., each value have the same level of certainty. If many measure can be assessed under the normality assumption, this is usually not the case in epidemiology with strictly positives rates and count values that 1) does not fit the normal distribution and 2) does not provide with the same degree of certainty since variances likely differ between district due to different population size, i.e., some district have very sparse data (with high variance) while other have adequate data (with lower variance).\n\n# dataset statistics\nm_cases <- mean(district$incidence)\nsd_cases <- sd(district$incidence)\n\nhist(district$incidence, probability = TRUE, ylim = c(0, 0.4), xlim = c(-5, 16), xlab = \"Number of cases\", ylab = \"Probability\", main = \"Histogram of observed incidence compared\\nto Normal and Poisson distributions\")\ncurve(dnorm(x, m_cases, sd_cases),col = \"blue\", lwd = 1, add = TRUE)\npoints(0:max(district$incidence), dpois(0:max(district$incidence), m_cases),type = 'b', pch = 20, col = \"red\", ylim = c(0, 0.6), lty = 2)\n\nlegend(\"topright\", legend = c(\"Normal distribution\", \"Poisson distribution\", \"Observed distribution\"), col = c(\"blue\", \"red\", \"black\"),pch = c(NA, 20, NA), lty = c(1, 2, 1))\n\n\n\n\nIn this tutorial, we used the Poisson distribution in our statistical tests.\n\n\n\n\n6.2.2 Test for spatial autocorrelation (Moran’s I test)\n\n6.2.2.1 The global Moran’s I test\nA popular test for spatial autocorrelation is the Moran’s test. This test tells us whether nearby units tend to exhibit similar incidences. It ranges from -1 to +1. A value of -1 denote that units with low rates are located near other units with high rates, while a Moran’s I value of +1 indicates a concentration of spatial units exhibiting similar rates.\n\n\n\n\n\n\nMoran’s I test\n\n\n\nThe Moran’s statistics is:\n\\[I = \\frac{N}{\\sum_{i=1}^N\\sum_{j=1}^Nw_{ij}}\\frac{\\sum_{i=1}^N\\sum_{j=1}^Nw_{ij}(Y_i-\\bar{Y})(Y_j - \\bar{Y})}{\\sum_{i=1}^N(Y_i-\\bar{Y})^2}\\] with:\n\n\\(N\\): the number of polygons,\n\\(w_{ij}\\): is a matrix of spatial weight with zeroes on the diagonal (i.e., \\(w_{ii}=0\\)). For example, if polygons are neighbors, the weight takes the value \\(1\\) otherwise it takes the value \\(0\\).\n\\(Y_i\\): the variable of interest,\n\\(\\bar{Y}\\): the mean value of \\(Y\\).\n\nUnder the Moran’s test, the statistics hypotheses are:\n\nH0: the distribution of cases is spatially independent, i.e., \\(I=0\\).\nH1: the distribution of cases is spatially autocorrelated, i.e., \\(I\\ne0\\).\n\n\n\nWe will compute the Moran’s statistics using spdep(R. Bivand et al. 2015) and Dcluster(Gómez-Rubio et al. 2015) packages. spdep package provides a collection of functions to analyze spatial correlations of polygons and works with sp objects. In this example, we use poly2nb() and nb2listw(). These functions respectively detect the neighboring polygons and assign weight corresponding to \\(1/\\#\\ of\\ neighbors\\). Dcluster package provides a set of functions for the detection of spatial clusters of disease using count data.\n\nlibrary(spdep) # Functions for creating spatial weight, spatial analysis\nlibrary(DCluster) # Package with functions for spatial cluster analysis\n\nqueen_nb <- poly2nb(district) # Neighbors according to queen case\nq_listw <- nb2listw(queen_nb, style = 'W') # row-standardized weights\n\n# Moran's I test\nm_test <- moranI.test(cases ~ offset(log(expected)), \n data = district,\n model = 'poisson',\n R = 499,\n listw = q_listw,\n n = length(district$cases), # number of regions\n S0 = Szero(q_listw)) # Global sum of weights\nprint(m_test)\n\nMoran's I test of spatial autocorrelation \n\n Type of boots.: parametric \n Model used when sampling: Poisson \n Number of simulations: 499 \n Statistic: 0.1566449 \n p-value : 0.014 \n\nplot(m_test)\n\n\n\n\nThe Moran’s statistics is here \\(I =\\) 0.16. When comparing its value to the H0 distribution (built under 499 simulations), the probability of observing such a I value under the null hypothesis, i.e. the distribution of cases is spatially independent, is \\(p_{value} =\\) 0.014. We therefore reject H0 with error risk of \\(\\alpha = 5\\%\\). The distribution of cases is therefore autocorrelated across districts in Cambodia.\n\n\n6.2.2.2 The Local Moran’s I LISA test\nThe global Moran’s test provides us a global statistical value informing whether autocorrelation occurs over the territory but does not inform on where does these correlations occurs, i.e., what is the locations of the clusters. To identify such cluster, we can decompose the Moran’s I statistic to extract local information of the level of correlation of each district and its neighbors. This is called the Local Moran’s I LISA statistic. Because the Local Moran’s I LISA statistic test each district for autocorrelation independently, concern is raised about multiple testing limitations that increase the Type I error (\\(\\alpha\\)) of the statistical tests. The use of local test should therefore be study in light of explore and describes clusters once the global test has detected autocorrelation.\n\n\n\n\n\n\nStatistical test\n\n\n\nFor each district \\(i\\), the Local Moran’s I statistics is:\n\\[I_i = \\frac{(Y_i-\\bar{Y})}{\\sum_{i=1}^N(Y_i-\\bar{Y})^2}\\sum_{j=1}^Nw_{ij}(Y_j - \\bar{Y}) \\text{ with } I = \\sum_{i=1}^NI_i/N\\]\n\n\nThe localmoran()function from the package spdep treats the variable of interest as if it was normally distributed. In some cases, this assumption could be reasonable for incidence rate, especially when the areal units of analysis have sufficiently large population count suggesting that the values have similar level of variances. Unfortunately, the local Moran’s test has not been implemented for Poisson distribution (population not large enough in some districts) in spdep package. However, Bivand et al. (R. S. Bivand et al. 2008) provided some code to manually perform the analysis using Poisson distribution and this code was further implemented in the course “Spatial Epidemiologyâ€.\n\n# Step 1 - Create the standardized deviation of observed from expected\nsd_lm <- (district$cases - district$expected) / sqrt(district$expected)\n\n# Step 2 - Create a spatially lagged version of standardized deviation of neighbors\nwsd_lm <- lag.listw(q_listw, sd_lm)\n\n# Step 3 - the local Moran's I is the product of step 1 and step 2\ndistrict$I_lm <- sd_lm * wsd_lm\n\n# Step 4 - setup parameters for simulation of the null distribution\n\n# Specify number of simulations to run\nnsim <- 499\n\n# Specify dimensions of result based on number of regions\nN <- length(district$expected)\n\n# Create a matrix of zeros to hold results, with a row for each county, and a column for each simulation\nsims <- matrix(0, ncol = nsim, nrow = N)\n\n# Step 5 - Start a for-loop to iterate over simulation columns\nfor(i in 1:nsim){\n y <- rpois(N, lambda = district$expected) # generate a random event count, given expected\n sd_lmi <- (y - district$expected) / sqrt(district$expected) # standardized local measure\n wsd_lmi <- lag.listw(q_listw, sd_lmi) # standardized spatially lagged measure\n sims[, i] <- sd_lmi * wsd_lmi # this is the I(i) statistic under this iteration of null\n}\n\n# Step 6 - For each county, test where the observed value ranks with respect to the null simulations\nxrank <- apply(cbind(district$I_lm, sims), 1, function(x) rank(x)[1])\n\n# Step 7 - Calculate the difference between observed rank and total possible (nsim)\ndiff <- nsim - xrank\ndiff <- ifelse(diff > 0, diff, 0)\n\n# Step 8 - Assuming a uniform distribution of ranks, calculate p-value for observed\n# given the null distribution generate from simulations\ndistrict$pval_lm <- punif((diff + 1) / (nsim + 1))\n\nBriefly, the process consist on 1) computing the I statistics for the observed data, 2) estimating the null distribution of the I statistics by performing random sampling into a poisson distribution and 3) comparing the observed I statistic with the null distribution to determine the probability to observe such value if the number of cases were spatially independent. For each district, we obtain a p-value based on the comparison of the observed value and the null distribution.\nA conventional way of plotting these results is to classify the districts into 5 classes based on local Moran’s I output. The classification of cluster that are significantly autocorrelated to their neighbors is performed based on a comparison of the scaled incidence in the district compared to the scaled weighted averaged incidence of it neighboring districts (computed with lag.listw()):\n\nDistricts that have higher-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local \\(I_i\\) statistic are defined as High-High (hotspot of the disease)\nDistricts that have lower-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local \\(I_i\\) statistic are defined as Low-Low (cold spot of the disease).\nDistricts that have higher-than-average rates in the index regions and lower-than-average rates in their neighbors, and showing statistically significant negative values for the local \\(I_i\\) statistic are defined as High-Low(outlier with high incidence in an area with low incidence).\nDistricts that have lower-than-average rates in the index regions and higher-than-average rates in their neighbors, and showing statistically significant negative values for the local \\(I_i\\) statistic are defined as Low-High (outlier of low incidence in area with high incidence).\nDistricts with non-significant values for the \\(I_i\\) statistic are defined as Non-significant.\n\n\n# create lagged local raw_rate - in other words the average of the queen neighbors value\n# values are scaled (centered and reduced) to be compared to average\ndistrict$lag_std <- scale(lag.listw(q_listw, var = district$incidence))\ndistrict$incidence_std <- scale(district$incidence)\n\n# extract pvalues\n# district$lm_pv <- lm_test[,5]\n\n# Classify local moran's outputs\ndistrict$lm_class <- NA\ndistrict$lm_class[district$incidence_std >=0 & district$lag_std >=0] <- 'High-High'\ndistrict$lm_class[district$incidence_std <=0 & district$lag_std <=0] <- 'Low-Low'\ndistrict$lm_class[district$incidence_std <=0 & district$lag_std >=0] <- 'Low-High'\ndistrict$lm_class[district$incidence_std >=0 & district$lag_std <=0] <- 'High-Low'\ndistrict$lm_class[district$pval_lm >= 0.05] <- 'Non-significant'\n\ndistrict$lm_class <- factor(district$lm_class, levels=c(\"High-High\", \"Low-Low\", \"High-Low\", \"Low-High\", \"Non-significant\") )\n\n# create map\nmf_map(x = district,\n var = \"lm_class\",\n type = \"typo\",\n cex = 2,\n col_na = \"white\",\n #val_order = c(\"High-High\", \"Low-Low\", \"High-Low\", \"Low-High\", \"Non-significant\") ,\n pal = c(\"#6D0026\" , \"blue\", \"white\") , # \"#FF755F\",\"#7FABD3\" ,\n leg_title = \"Clusters\")\n\nmf_layout(title = \"Cluster using Local Moran's I statistic\")\n\n\n\n\n\n\n\n6.2.3 Spatial scan statistics\nWhile Moran’s indices focus on testing for autocorrelation between neighboring polygons (under the null assumption of spatial independence), the spatial scan statistic aims at identifying an abnormal higher risk in a given region compared to the risk outside of this region (under the null assumption of homogeneous distribution). The conception of a cluster is therefore different between the two methods.\nThe function kulldorff from the package SpatialEpi (Kim and Wakefield 2010) is a simple tool to implement spatial-only scan statistics.\n\n\n\n\n\n\nKulldorf test\n\n\n\nUnder the kulldorff test, the statistics hypotheses are:\n\nH0: the risk is constant over the area, i.e., there is a spatial homogeneity of the incidence.\nH1: a particular window have higher incidence than the rest of the area , i.e., there is a spatial heterogeneity of incidence.\n\n\n\nBriefly, the kulldorff scan statistics scan the area for clusters using several steps:\n\nIt create a circular window of observation by defining a single location and an associated radius of the windows varying from 0 to a large number that depends on population distribution (largest radius could include 50% of the population).\nIt aggregates the count of events and the population at risk (or an expected count of events) inside and outside the window of observation.\nFinally, it computes the likelihood ratio and test whether the risk is equal inside versus outside the windows (H0) or greater inside the observed window (H1). The H0 distribution is estimated by simulating the distribution of counts under the null hypothesis (homogeneous risk).\nThese 3 steps are repeated for each location and each possible windows-radii.\n\nWhile we test the significance of a large number of observation windows, one can raise concern about multiple testing and Type I error. This approach however suggest that we are not interest in a set of signifiant cluster but only in a most-likely cluster. This a priori restriction eliminate concern for multpile comparison since the test is simplified to a statistically significance of one single most-likely cluster.\nBecause we tested all-possible locations and window-radius, we can also choose to look at secondary clusters. In this case, you should keep in mind that increasing the number of secondary cluster you select, increases the risk for Type I error.\n\nlibrary(\"SpatialEpi\")\n\nThe use of R spatial object is not implements in kulldorff() function. It uses instead matrix of xy coordinates that represents the centroids of the districts. A given district is included into the observed circular window if its centroids fall into the circle.\n\ndistrict_xy <- st_centroid(district) %>% \n st_coordinates()\n\nhead(district_xy)\n\n X Y\n1 330823.3 1464560\n2 749758.3 1541787\n3 468384.0 1277007\n4 494548.2 1215261\n5 459644.2 1194615\n6 360528.3 1516339\n\n\nWe can then call kulldorff function (you are strongly encouraged to call ?kulldorff to properly call the function). The alpha.level threshold filter for the secondary clusters that will be retained. The most-likely cluster will be saved whatever its significance.\n\nkd_Wfever <- kulldorff(district_xy, \n cases = district$cases,\n population = district$T_POP,\n expected.cases = district$expected,\n pop.upper.bound = 0.5, # include maximum 50% of the population in a windows\n n.simulations = 499,\n alpha.level = 0.2)\n\n\n\n\nThe function plot the histogram of the distribution of log-likelihood ratio simulated under the null hypothesis that is estimated based on Monte Carlo simulations. The observed value of the most significant cluster identified from all possible scans is compared to the distribution to determine significance. All outputs are saved into an R object, here called kd_Wfever. Unfortunately, the package did not develop any summary and visualization of the results but we can explore the output object.\n\nnames(kd_Wfever)\n\n[1] \"most.likely.cluster\" \"secondary.clusters\" \"type\" \n[4] \"log.lkhd\" \"simulated.log.lkhd\" \n\n\nFirst, we can focus on the most likely cluster and explore its characteristics.\n\n# We can see which districts (r number) belong to this cluster\nkd_Wfever$most.likely.cluster$location.IDs.included\n\n [1] 48 93 66 180 133 29 194 118 50 144 31 141 3 117 22 43 142\n\n# standardized incidence ratio\nkd_Wfever$most.likely.cluster$SMR\n\n[1] 2.303106\n\n# number of observed and expected cases in this cluster\nkd_Wfever$most.likely.cluster$number.of.cases\n\n[1] 122\n\nkd_Wfever$most.likely.cluster$expected.cases\n\n[1] 52.97195\n\n\n17 districts belong to the cluster and its number of cases is 2.3 times higher than the expected number of cases.\nSimilarly, we could study the secondary clusters. Results are saved in a list.\n\n# We can see which districts (r number) belong to this cluster\nlength(kd_Wfever$secondary.clusters)\n\n[1] 1\n\n# retrieve data for all secondary clusters into a table\ndf_secondary_clusters <- data.frame(SMR = sapply(kd_Wfever$secondary.clusters, '[[', 5), \n number.of.cases = sapply(kd_Wfever$secondary.clusters, '[[', 3),\n expected.cases = sapply(kd_Wfever$secondary.clusters, '[[', 4),\n p.value = sapply(kd_Wfever$secondary.clusters, '[[', 8))\n\nprint(df_secondary_clusters)\n\n SMR number.of.cases expected.cases p.value\n1 3.767698 16 4.246625 0.014\n\n\nWe only have one secondary cluster composed of one district.\n\n# create empty column to store cluster informations\ndistrict$k_cluster <- NA\n\n# save cluster information from kulldorff outputs\ndistrict$k_cluster[kd_Wfever$most.likely.cluster$location.IDs.included] <- 'Most likely cluster'\n\nfor(i in 1:length(kd_Wfever$secondary.clusters)){\ndistrict$k_cluster[kd_Wfever$secondary.clusters[[i]]$location.IDs.included] <- paste(\n 'Secondary cluster', i, sep = '')\n}\n\n#district$k_cluster[is.na(district$k_cluster)] <- \"No cluster\"\n\n\n# create map\nmf_map(x = district,\n var = \"k_cluster\",\n type = \"typo\",\n cex = 2,\n col_na = \"white\",\n pal = mf_get_pal(palette = \"Reds\", n = 3)[1:2],\n leg_title = \"Clusters\")\n\nmf_layout(title = \"Cluster using kulldorf scan statistic\")\n\n\n\n\n\n\n\n\n\n\nTo go further …\n\n\n\nIn this example, the expected number of cases was defined using the population count but note that standardization over other variables as age could also be implemented with the strata parameter in the kulldorff() function.\nIn addition, this cluster analysis was performed solely using the spatial scan but you should keep in mind that this method of cluster detection can be implemented for spatio-temporal data as well where the cluster definition is an abnormal number of cases in a delimited spatial area and during a given period of time. The windows of observation are therefore defined for a different center, radius and time-period. You should look at the function scan_ep_poisson() function in the package scanstatistic (Allévius 2018) for this analysis.\n\n\n\n\n\n\nAllévius, Benjamin. 2018. “Scanstatistics: Space-Time Anomaly Detection Using Scan Statistics.†Journal of Open Source Software 3 (25): 515.\n\n\nBivand, Roger S, Edzer J Pebesma, Virgilio Gómez-Rubio, and Edzer Jan Pebesma. 2008. Applied Spatial Data Analysis with r. Vol. 747248717. Springer.\n\n\nBivand, Roger, Micah Altman, Luc Anselin, Renato Assunção, Olaf Berke, Andrew Bernat, and Guillaume Blanchet. 2015. “Package ‘Spdep’.†The Comprehensive R Archive Network.\n\n\nGómez-Rubio, Virgilio, Juan Ferrándiz-Ferragud, Antonio López-QuıÌlez, et al. 2015. “Package ‘DCluster’.â€\n\n\nKim, Albert Y, and Jon Wakefield. 2010. “R Data and Methods for Spatial Epidemiology: The SpatialEpi Package.†Dept of Statistics, University of Washington."
+ "text": "6.2 Cluster analysis\n\n6.2.1 General introduction\nWhy studying clusters in epidemiology? Cluster analysis help identifying unusual patterns that occurs during a given period of time. The underlying ultimate goal of such analysis is to explain the observation of such patterns. In epidemiology, we can distinguish two types of process that would explain heterogeneity in case distribution:\n\nThe 1st order effects are the spatial variations of cases distribution caused by underlying properties of environment or the population structure itself. In such process individual get infected independently from the rest of the population. Such process includes the infection through an environment at risk as, for example, air pollution, contaminated waters or soils and UV exposition. This effect assume that the observed pattern is caused by a difference in risk intensity.\nThe 2nd order effects describes process of spread, contagion and diffusion of diseases caused by interactions between individuals. This includes transmission of infectious disease by proximity, but also the transmission of non-infectious disease, for example, with the diffusion of social norms within networks. This effect assume that the observed pattern is caused by correlations or co-variations.\n\nNo statistical methods could distinguish between these competing processes since their outcome results in similar pattern of points. The cluster analysis help describing the magnitude and the location of pattern but in no way could answer the question of why such patterns occurs. It is therefore a step that help detecting cluster for description and surveillance purpose and rising hypothesis on the underlying process that will lead further investigations.\nKnowledge about the disease and its transmission process could orientate the choice of the methods of study. We presented in this brief tutorial two methods of cluster detection, the Moran’s I test that test for spatial independence (likely related to 2nd order effects) and the scan statistics that test for homogeneous distribution (likely related 1st order effects). It relies on epidemiologist to select the tools that best serve the studied question.\n\n\n\n\n\n\nStatistic tests and distributions\n\n\n\nIn statistics, problems are usually expressed by defining two hypotheses: the null hypothesis (H0), i.e., an a priori hypothesis of the studied phenomenon (e.g., the situation is a random) and the alternative hypothesis (H1), e.g., the situation is not random. The main principle is to measure how likely the observed situation belong to the ensemble of situation that are possible under the H0 hypothesis.\nIn mathematics, a probability distribution is a mathematical expression that represents what we would expect due to random chance. The choice of the probability distribution relies on the type of data you use (continuous, count, binary). In general, three distribution a used while studying disease rates, the Binomial, the Poisson and the Poisson-gamma mixture (also known as negative binomial) distributions.\nMany the statistical tests assume by default that data are normally distributed. It implies that your variable is continuous and that all data could easily be represented by two parameters, the mean and the variance, i.e., each value have the same level of certainty. If many measure can be assessed under the normality assumption, this is usually not the case in epidemiology with strictly positives rates and count values that 1) does not fit the normal distribution and 2) does not provide with the same degree of certainty since variances likely differ between district due to different population size, i.e., some district have very sparse data (with high variance) while other have adequate data (with lower variance).\n\n# dataset statistics\nm_cases <- mean(district$incidence)\nsd_cases <- sd(district$incidence)\n\nhist(district$incidence, probability = TRUE, ylim = c(0, 0.4), xlim = c(-5, 16), xlab = \"Number of cases\", ylab = \"Probability\", main = \"Histogram of observed incidence compared\\nto Normal and Poisson distributions\")\ncurve(dnorm(x, m_cases, sd_cases),col = \"blue\", lwd = 1, add = TRUE)\npoints(0:max(district$incidence), dpois(0:max(district$incidence), m_cases),type = 'b', pch = 20, col = \"red\", ylim = c(0, 0.6), lty = 2)\n\nlegend(\"topright\", legend = c(\"Normal distribution\", \"Poisson distribution\", \"Observed distribution\"), col = c(\"blue\", \"red\", \"black\"),pch = c(NA, 20, NA), lty = c(1, 2, 1))\n\n\n\n\nIn this tutorial, we used the Poisson distribution in our statistical tests.\n\n\n\n\n6.2.2 Test for spatial autocorrelation (Moran’s I test)\n\n6.2.2.1 The global Moran’s I test\nA popular test for spatial autocorrelation is the Moran’s test. This test tells us whether nearby units tend to exhibit similar incidences. It ranges from -1 to +1. A value of -1 denote that units with low rates are located near other units with high rates, while a Moran’s I value of +1 indicates a concentration of spatial units exhibiting similar rates.\n\n\n\n\n\n\nMoran’s I test\n\n\n\nThe Moran’s statistics is:\n\\[I = \\frac{N}{\\sum_{i=1}^N\\sum_{j=1}^Nw_{ij}}\\frac{\\sum_{i=1}^N\\sum_{j=1}^Nw_{ij}(Y_i-\\bar{Y})(Y_j - \\bar{Y})}{\\sum_{i=1}^N(Y_i-\\bar{Y})^2}\\] with:\n\n\\(N\\): the number of polygons,\n\\(w_{ij}\\): is a matrix of spatial weight with zeroes on the diagonal (i.e., \\(w_{ii}=0\\)). For example, if polygons are neighbors, the weight takes the value \\(1\\) otherwise it takes the value \\(0\\).\n\\(Y_i\\): the variable of interest,\n\\(\\bar{Y}\\): the mean value of \\(Y\\).\n\nUnder the Moran’s test, the statistics hypotheses are:\n\nH0: the distribution of cases is spatially independent, i.e., \\(I=0\\).\nH1: the distribution of cases is spatially autocorrelated, i.e., \\(I\\ne0\\).\n\n\n\nWe will compute the Moran’s statistics using spdep(R. Bivand et al. 2015) and Dcluster(Gómez-Rubio et al. 2015) packages. spdep package provides a collection of functions to analyze spatial correlations of polygons and works with sp objects. In this example, we use poly2nb() and nb2listw(). These functions respectively detect the neighboring polygons and assign weight corresponding to \\(1/\\#\\ of\\ neighbors\\). Dcluster package provides a set of functions for the detection of spatial clusters of disease using count data.\n\n#install.packages(\"spdep\")\n#install.packages(\"DCluster\")\nlibrary(spdep) # Functions for creating spatial weight, spatial analysis\nlibrary(DCluster) # Package with functions for spatial cluster analysis\n\nqueen_nb <- poly2nb(district) # Neighbors according to queen case\nq_listw <- nb2listw(queen_nb, style = 'W') # row-standardized weights\n\n# Moran's I test\nm_test <- moranI.test(cases ~ offset(log(expected)), \n data = district,\n model = 'poisson',\n R = 499,\n listw = q_listw,\n n = length(district$cases), # number of regions\n S0 = Szero(q_listw)) # Global sum of weights\nprint(m_test)\n\nMoran's I test of spatial autocorrelation \n\n Type of boots.: parametric \n Model used when sampling: Poisson \n Number of simulations: 499 \n Statistic: 0.1566449 \n p-value : 0.008 \n\nplot(m_test)\n\n\n\n\nThe Moran’s statistics is here \\(I =\\) 0.16. When comparing its value to the H0 distribution (built under 499 simulations), the probability of observing such a I value under the null hypothesis, i.e. the distribution of cases is spatially independent, is \\(p_{value} =\\) 0.008. We therefore reject H0 with error risk of \\(\\alpha = 5\\%\\). The distribution of cases is therefore autocorrelated across districts in Cambodia.\n\n\n6.2.2.2 The Local Moran’s I LISA test\nThe global Moran’s test provides us a global statistical value informing whether autocorrelation occurs over the territory but does not inform on where does these correlations occurs, i.e., what is the locations of the clusters. To identify such cluster, we can decompose the Moran’s I statistic to extract local information of the level of correlation of each district and its neighbors. This is called the Local Moran’s I LISA statistic. Because the Local Moran’s I LISA statistic test each district for autocorrelation independently, concern is raised about multiple testing limitations that increase the Type I error (\\(\\alpha\\)) of the statistical tests. The use of local test should therefore be study in light of explore and describes clusters once the global test has detected autocorrelation.\n\n\n\n\n\n\nStatistical test\n\n\n\nFor each district \\(i\\), the Local Moran’s I statistics is:\n\\[I_i = \\frac{(Y_i-\\bar{Y})}{\\sum_{i=1}^N(Y_i-\\bar{Y})^2}\\sum_{j=1}^Nw_{ij}(Y_j - \\bar{Y}) \\text{ with } I = \\sum_{i=1}^NI_i/N\\]\n\n\nThe localmoran()function from the package spdep treats the variable of interest as if it was normally distributed. In some cases, this assumption could be reasonable for incidence rate, especially when the areal units of analysis have sufficiently large population count suggesting that the values have similar level of variances. Unfortunately, the local Moran’s test has not been implemented for Poisson distribution (population not large enough in some districts) in spdep package. However, Bivand et al. (R. S. Bivand et al. 2008) provided some code to manually perform the analysis using Poisson distribution and this code was further implemented in the course “Spatial Epidemiologyâ€.\n\n# Step 1 - Create the standardized deviation of observed from expected\nsd_lm <- (district$cases - district$expected) / sqrt(district$expected)\n\n# Step 2 - Create a spatially lagged version of standardized deviation of neighbors\nwsd_lm <- lag.listw(q_listw, sd_lm)\n\n# Step 3 - the local Moran's I is the product of step 1 and step 2\ndistrict$I_lm <- sd_lm * wsd_lm\n\n# Step 4 - setup parameters for simulation of the null distribution\n\n# Specify number of simulations to run\nnsim <- 499\n\n# Specify dimensions of result based on number of regions\nN <- length(district$expected)\n\n# Create a matrix of zeros to hold results, with a row for each county, and a column for each simulation\nsims <- matrix(0, ncol = nsim, nrow = N)\n\n# Step 5 - Start a for-loop to iterate over simulation columns\nfor(i in 1:nsim){\n y <- rpois(N, lambda = district$expected) # generate a random event count, given expected\n sd_lmi <- (y - district$expected) / sqrt(district$expected) # standardized local measure\n wsd_lmi <- lag.listw(q_listw, sd_lmi) # standardized spatially lagged measure\n sims[, i] <- sd_lmi * wsd_lmi # this is the I(i) statistic under this iteration of null\n}\n\n# Step 6 - For each county, test where the observed value ranks with respect to the null simulations\nxrank <- apply(cbind(district$I_lm, sims), 1, function(x) rank(x)[1])\n\n# Step 7 - Calculate the difference between observed rank and total possible (nsim)\ndiff <- nsim - xrank\ndiff <- ifelse(diff > 0, diff, 0)\n\n# Step 8 - Assuming a uniform distribution of ranks, calculate p-value for observed\n# given the null distribution generate from simulations\ndistrict$pval_lm <- punif((diff + 1) / (nsim + 1))\n\nBriefly, the process consist on 1) computing the I statistics for the observed data, 2) estimating the null distribution of the I statistics by performing random sampling into a poisson distribution and 3) comparing the observed I statistic with the null distribution to determine the probability to observe such value if the number of cases were spatially independent. For each district, we obtain a p-value based on the comparison of the observed value and the null distribution.\nA conventional way of plotting these results is to classify the districts into 5 classes based on local Moran’s I output. The classification of cluster that are significantly autocorrelated to their neighbors is performed based on a comparison of the scaled incidence in the district compared to the scaled weighted averaged incidence of it neighboring districts (computed with lag.listw()):\n\nDistricts that have higher-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local \\(I_i\\) statistic are defined as High-High (hotspot of the disease)\nDistricts that have lower-than-average rates in both index regions and their neighbors and showing statistically significant positive values for the local \\(I_i\\) statistic are defined as Low-Low (cold spot of the disease).\nDistricts that have higher-than-average rates in the index regions and lower-than-average rates in their neighbors, and showing statistically significant negative values for the local \\(I_i\\) statistic are defined as High-Low(outlier with high incidence in an area with low incidence).\nDistricts that have lower-than-average rates in the index regions and higher-than-average rates in their neighbors, and showing statistically significant negative values for the local \\(I_i\\) statistic are defined as Low-High (outlier of low incidence in area with high incidence).\nDistricts with non-significant values for the \\(I_i\\) statistic are defined as Non-significant.\n\n\n# create lagged local raw_rate - in other words the average of the queen neighbors value\n# values are scaled (centered and reduced) to be compared to average\ndistrict$lag_std <- scale(lag.listw(q_listw, var = district$incidence))\ndistrict$incidence_std <- scale(district$incidence)\n\n# extract pvalues\n# district$lm_pv <- lm_test[,5]\n\n# Classify local moran's outputs\ndistrict$lm_class <- NA\ndistrict$lm_class[district$incidence_std >=0 & district$lag_std >=0] <- 'High-High'\ndistrict$lm_class[district$incidence_std <=0 & district$lag_std <=0] <- 'Low-Low'\ndistrict$lm_class[district$incidence_std <=0 & district$lag_std >=0] <- 'Low-High'\ndistrict$lm_class[district$incidence_std >=0 & district$lag_std <=0] <- 'High-Low'\ndistrict$lm_class[district$pval_lm >= 0.05] <- 'Non-significant'\n\ndistrict$lm_class <- factor(district$lm_class, levels=c(\"High-High\", \"Low-Low\", \"High-Low\", \"Low-High\", \"Non-significant\") )\n\n# create map\nmf_map(x = district,\n var = \"lm_class\",\n type = \"typo\",\n cex = 2,\n col_na = \"white\",\n #val_order = c(\"High-High\", \"Low-Low\", \"High-Low\", \"Low-High\", \"Non-significant\") ,\n pal = c(\"#6D0026\" , \"blue\", \"white\") , # \"#FF755F\",\"#7FABD3\" ,\n leg_title = \"Clusters\")\n\nmf_layout(title = \"Cluster using Local Moran's I statistic\")\n\n\n\n\n\n\n\n6.2.3 Spatial scan statistics\nWhile Moran’s indices focus on testing for autocorrelation between neighboring polygons (under the null assumption of spatial independence), the spatial scan statistic aims at identifying an abnormal higher risk in a given region compared to the risk outside of this region (under the null assumption of homogeneous distribution). The conception of a cluster is therefore different between the two methods.\nThe function kulldorff from the package SpatialEpi (Kim and Wakefield 2010) is a simple tool to implement spatial-only scan statistics.\n\n\n\n\n\n\nKulldorf test\n\n\n\nUnder the kulldorff test, the statistics hypotheses are:\n\nH0: the risk is constant over the area, i.e., there is a spatial homogeneity of the incidence.\nH1: a particular window have higher incidence than the rest of the area , i.e., there is a spatial heterogeneity of incidence.\n\n\n\nBriefly, the kulldorff scan statistics scan the area for clusters using several steps:\n\nIt create a circular window of observation by defining a single location and an associated radius of the windows varying from 0 to a large number that depends on population distribution (largest radius could include 50% of the population).\nIt aggregates the count of events and the population at risk (or an expected count of events) inside and outside the window of observation.\nFinally, it computes the likelihood ratio and test whether the risk is equal inside versus outside the windows (H0) or greater inside the observed window (H1). The H0 distribution is estimated by simulating the distribution of counts under the null hypothesis (homogeneous risk).\nThese 3 steps are repeated for each location and each possible windows-radii.\n\nWhile we test the significance of a large number of observation windows, one can raise concern about multiple testing and Type I error. This approach however suggest that we are not interest in a set of signifiant cluster but only in a most-likely cluster. This a priori restriction eliminate concern for multpile comparison since the test is simplified to a statistically significance of one single most-likely cluster.\nBecause we tested all-possible locations and window-radius, we can also choose to look at secondary clusters. In this case, you should keep in mind that increasing the number of secondary cluster you select, increases the risk for Type I error.\n\n#install.packages(\"SpatialEpi\")\nlibrary(\"SpatialEpi\")\n\nThe use of R spatial object is not implements in kulldorff() function. It uses instead matrix of xy coordinates that represents the centroids of the districts. A given district is included into the observed circular window if its centroids fall into the circle.\n\ndistrict_xy <- st_centroid(district) %>% \n st_coordinates()\n\nhead(district_xy)\n\n X Y\n1 330823.3 1464560\n2 749758.3 1541787\n3 468384.0 1277007\n4 494548.2 1215261\n5 459644.2 1194615\n6 360528.3 1516339\n\n\nWe can then call kulldorff function (you are strongly encouraged to call ?kulldorff to properly call the function). The alpha.level threshold filter for the secondary clusters that will be retained. The most-likely cluster will be saved whatever its significance.\n\nkd_Wfever <- kulldorff(district_xy, \n cases = district$cases,\n population = district$T_POP,\n expected.cases = district$expected,\n pop.upper.bound = 0.5, # include maximum 50% of the population in a windows\n n.simulations = 499,\n alpha.level = 0.2)\n\n\n\n\nThe function plot the histogram of the distribution of log-likelihood ratio simulated under the null hypothesis that is estimated based on Monte Carlo simulations. The observed value of the most significant cluster identified from all possible scans is compared to the distribution to determine significance. All outputs are saved into an R object, here called kd_Wfever. Unfortunately, the package did not develop any summary and visualization of the results but we can explore the output object.\n\nnames(kd_Wfever)\n\n[1] \"most.likely.cluster\" \"secondary.clusters\" \"type\" \n[4] \"log.lkhd\" \"simulated.log.lkhd\" \n\n\nFirst, we can focus on the most likely cluster and explore its characteristics.\n\n# We can see which districts (r number) belong to this cluster\nkd_Wfever$most.likely.cluster$location.IDs.included\n\n [1] 48 93 66 180 133 29 194 118 50 144 31 141 3 117 22 43 142\n\n# standardized incidence ratio\nkd_Wfever$most.likely.cluster$SMR\n\n[1] 2.303106\n\n# number of observed and expected cases in this cluster\nkd_Wfever$most.likely.cluster$number.of.cases\n\n[1] 122\n\nkd_Wfever$most.likely.cluster$expected.cases\n\n[1] 52.97195\n\n\n17 districts belong to the cluster and its number of cases is 2.3 times higher than the expected number of cases.\nSimilarly, we could study the secondary clusters. Results are saved in a list.\n\n# We can see which districts (r number) belong to this cluster\nlength(kd_Wfever$secondary.clusters)\n\n[1] 1\n\n# retrieve data for all secondary clusters into a table\ndf_secondary_clusters <- data.frame(SMR = sapply(kd_Wfever$secondary.clusters, '[[', 5), \n number.of.cases = sapply(kd_Wfever$secondary.clusters, '[[', 3),\n expected.cases = sapply(kd_Wfever$secondary.clusters, '[[', 4),\n p.value = sapply(kd_Wfever$secondary.clusters, '[[', 8))\n\nprint(df_secondary_clusters)\n\n SMR number.of.cases expected.cases p.value\n1 3.767698 16 4.246625 0.008\n\n\nWe only have one secondary cluster composed of one district.\n\n# create empty column to store cluster informations\ndistrict$k_cluster <- NA\n\n# save cluster information from kulldorff outputs\ndistrict$k_cluster[kd_Wfever$most.likely.cluster$location.IDs.included] <- 'Most likely cluster'\n\nfor(i in 1:length(kd_Wfever$secondary.clusters)){\ndistrict$k_cluster[kd_Wfever$secondary.clusters[[i]]$location.IDs.included] <- paste(\n 'Secondary cluster', i, sep = '')\n}\n\n#district$k_cluster[is.na(district$k_cluster)] <- \"No cluster\"\n\n\n# create map\nmf_map(x = district,\n var = \"k_cluster\",\n type = \"typo\",\n cex = 2,\n col_na = \"white\",\n pal = mf_get_pal(palette = \"Reds\", n = 3)[1:2],\n leg_title = \"Clusters\")\n\nmf_layout(title = \"Cluster using kulldorf scan statistic\")\n\n\n\n\n\n\n\n\n\n\nTo go further …\n\n\n\nIn this example, the expected number of cases was defined using the population count but note that standardization over other variables as age could also be implemented with the strata parameter in the kulldorff() function.\nIn addition, this cluster analysis was performed solely using the spatial scan but you should keep in mind that this method of cluster detection can be implemented for spatio-temporal data as well where the cluster definition is an abnormal number of cases in a delimited spatial area and during a given period of time. The windows of observation are therefore defined for a different center, radius and time-period. You should look at the function scan_ep_poisson() function in the package scanstatistic (Allévius 2018) for this analysis.\n\n\n\n\n\n\nAllévius, Benjamin. 2018. “Scanstatistics: Space-Time Anomaly Detection Using Scan Statistics.†Journal of Open Source Software 3 (25): 515.\n\n\nBivand, Roger S, Edzer J Pebesma, Virgilio Gómez-Rubio, and Edzer Jan Pebesma. 2008. Applied Spatial Data Analysis with r. Vol. 747248717. Springer.\n\n\nBivand, Roger, Micah Altman, Luc Anselin, Renato Assunção, Olaf Berke, Andrew Bernat, and Guillaume Blanchet. 2015. “Package ‘Spdep’.†The Comprehensive R Archive Network.\n\n\nGómez-Rubio, Virgilio, Juan Ferrándiz-Ferragud, Antonio López-QuıÌlez, et al. 2015. “Package ‘DCluster’.â€\n\n\nKim, Albert Y, and Jon Wakefield. 2010. “R Data and Methods for Spatial Epidemiology: The SpatialEpi Package.†Dept of Statistics, University of Washington."
},
{
"objectID": "01-introduction.html",
--
GitLab